Compare commits
2 Commits
1d9ece8182
...
06aded392e
| Author | SHA1 | Date | |
|---|---|---|---|
|
06aded392e | ||
|
|
3fb8566bc5 |
@ -449,3 +449,4 @@ remote: Processed 1 references in total
|
||||
To https://git.valavala.com/ai_member_only/ai_member_xiaoban
|
||||
2234f3e..0e9acbd master -> master
|
||||
[2026-05-28 08:10:17] 工作区备份成功：自动备份 2026-05-28 08:10:01
|
||||
[2026-05-29 08:10:01] 开始备份工作区...
|
||||
|
||||
@ -5,3 +5,8 @@
|
||||
{"type":"memory.recall.recorded","timestamp":"2026-05-27T13:30:03.421Z","query":"vala_knowledgebase 瓦拉星球 游戏剧情 世界观 角色","resultCount":3,"results":[{"path":"memory/2026-05-24.md","startLine":106,"endLine":126,"score":1},{"path":"memory/2026-05-24.md","startLine":85,"endLine":110,"score":1},{"path":"memory/2026-05-24.md","startLine":1,"endLine":30,"score":1}]}
|
||||
{"type":"memory.recall.recorded","timestamp":"2026-05-28T07:37:09.223Z","query":"李若松 Cris 近期对话 任务","resultCount":1,"results":[{"path":"memory/2026-05-13.md","startLine":1,"endLine":16,"score":1}]}
|
||||
{"type":"memory.recall.recorded","timestamp":"2026-05-28T09:07:57.953Z","query":"数据服务边界 xiaoban data boundary 查询输出 生产表","resultCount":5,"results":[{"path":"memory/2026-05-24.md","startLine":46,"endLine":71,"score":1},{"path":"memory/2026-05-24.md","startLine":66,"endLine":92,"score":1},{"path":"memory/2026-05-28.md","startLine":1,"endLine":33,"score":1},{"path":"memory/2026-05-24.md","startLine":106,"endLine":126,"score":1},{"path":"memory/2026-05-24.md","startLine":85,"endLine":110,"score":1}]}
|
||||
{"type":"memory.recall.recorded","timestamp":"2026-05-28T12:29:25.867Z","query":"订单详情 陈逸鸫 群聊","resultCount":2,"results":[{"path":"memory/2026-05-28.md","startLine":147,"endLine":179,"score":1},{"path":"memory/2026-05-28.md","startLine":313,"endLine":345,"score":1}]}
|
||||
{"type":"memory.recall.recorded","timestamp":"2026-05-28T12:30:30.129Z","query":"订单表 小龙 王虹茗 订单详情","resultCount":4,"results":[{"path":"memory/2026-05-24.md","startLine":46,"endLine":71,"score":1},{"path":"memory/2026-05-24.md","startLine":85,"endLine":110,"score":1},{"path":"memory/2026-05-28.md","startLine":147,"endLine":179,"score":1},{"path":"memory/2026-05-28.md","startLine":313,"endLine":345,"score":1}]}
|
||||
{"type":"memory.recall.recorded","timestamp":"2026-05-28T12:30:30.129Z","query":"订单明细 表结构 数据库 order","resultCount":6,"results":[{"path":"memory/2026-05-28.md","startLine":147,"endLine":179,"score":1},{"path":"memory/2026-05-28.md","startLine":313,"endLine":345,"score":1},{"path":"memory/2026-05-24.md","startLine":46,"endLine":71,"score":1},{"path":"memory/2026-05-24.md","startLine":85,"endLine":110,"score":1},{"path":"memory/2026-05-28.md","startLine":122,"endLine":152,"score":1},{"path":"memory/2026-05-28.md","startLine":288,"endLine":318,"score":1}]}
|
||||
{"type":"memory.recall.recorded","timestamp":"2026-05-28T20:51:03.908Z","query":"陈逸鸫 对话 任务 查询","resultCount":10,"results":[{"path":"memory/2026-05-13.md","startLine":1,"endLine":16,"score":1},{"path":"memory/2026-05-28.md","startLine":504,"endLine":536,"score":1},{"path":"memory/2026-05-28.md","startLine":337,"endLine":366,"score":1},{"path":"memory/2026-05-28.md","startLine":74,"endLine":100,"score":1},{"path":"memory/2026-05-28.md","startLine":240,"endLine":266,"score":1},{"path":"memory/2026-05-28.md","startLine":432,"endLine":458,"score":1},{"path":"memory/2026-05-28.md","startLine":530,"endLine":557,"score":1},{"path":"memory/2026-05-28.md","startLine":94,"endLine":127,"score":1},{"path":"memory/2026-05-28.md","startLine":260,"endLine":293,"score":1},{"path":"memory/2026-05-28.md","startLine":452,"endLine":485,"score":1}]}
|
||||
{"type":"memory.recall.recorded","timestamp":"2026-05-28T20:57:42.055Z","query":"xhs-tech-dashboard repo git clone url","resultCount":5,"results":[{"path":"memory/2026-05-28.md","startLine":94,"endLine":127,"score":1},{"path":"memory/2026-05-28.md","startLine":260,"endLine":293,"score":1},{"path":"memory/2026-05-28.md","startLine":452,"endLine":485,"score":1},{"path":"memory/2026-05-24.md","startLine":122,"endLine":132,"score":1},{"path":"memory/2026-05-28.md","startLine":122,"endLine":152,"score":1}]}
|
||||
|
||||
@ -1,6 +1,6 @@
|
||||
{
|
||||
"version": 1,
|
||||
"updatedAt": "2026-05-28T09:07:57.953Z",
|
||||
"updatedAt": "2026-05-28T20:57:42.055Z",
|
||||
"entries": {
|
||||
"memory:memory/2026-05-24.md:1:30": {
|
||||
"key": "memory:memory/2026-05-24.md:1:30",
|
||||
@ -141,17 +141,19 @@
|
||||
"endLine": 71,
|
||||
"source": "memory",
|
||||
"snippet": "### 增强: 报告开头加入角色基本信息 (2026-05-24) [刘庆逊提出] 在 studytime-analysis 输出中加入角色基本信息，包括： - 角色ID、账号ID、角色名字、性别、年龄、账号手机号后4位 **数据源（新增）**： - MySQL Online `vala_user` 库 - `vala_app_character` 表：id, account_id, nickname, gender(0=女/1=男), birthday(varchar \"YYYY-MM-DD\") - `vala_app_account` 表：id, tel(已脱敏如 186****1625) - 手机号已脱敏，直接取后4位；年龄从 birthday 计算 **修改文件**： - `skills/studytime-analysis/scripts/studytime_analysis.py`：新增 MySQL 连接函数 `get_mysql_connection()` 和 `fetch_role_info(role_id)`，更新 `format_report()` 输出基本角色信息 - 已验证 2895 正常运行输出 - 已同步 SkillHub + Git ### Unit 显示修复: 季度名称 → 全局单元编号 (2026-05-24) [刘庆逊提出] HTML 报告中 Unit 列显示错误——显示的是季度名称（如\"小镇时光\"\"钢铁之心\"）而非单元数字（0-48）。 **根因分析**： - `vala_game_chapter`（MySQL）无 `unit_",
|
||||
"recallCount": 3,
|
||||
"recallCount": 5,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 3,
|
||||
"totalScore": 5,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-25T05:47:41.388Z",
|
||||
"lastRecalledAt": "2026-05-28T09:07:57.953Z",
|
||||
"lastRecalledAt": "2026-05-28T12:30:30.129Z",
|
||||
"queryHashes": [
|
||||
"9aff8ec9594a",
|
||||
"566b5958861e",
|
||||
"c6c7ff4ed75d"
|
||||
"c6c7ff4ed75d",
|
||||
"7e2572c3140a",
|
||||
"4596e377d39b"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-25",
|
||||
@ -176,18 +178,20 @@
|
||||
"endLine": 110,
|
||||
"source": "memory",
|
||||
"snippet": "- `skills/studytime-analysis/scripts/studytime_analysis.py` — 重写 `fetch_chapter_info_map()`，新增全局 unit_index 计算；HTML 模板更新为 Level/Unit/Lesson 三列 - 已为角色 32009（zyl）重新生成 HTML 并发送 - 已同步 Git + SkillHub ## 新建技能: studycourse-analysis (2026-05-24) [刘庆逊提出] 创建角色上课情况分析技能，从四维度分析角色学习数据。 ### 技能结构 - `skills/studycourse-analysis/SKILL.md` — 技能定义 - `skills/studycourse-analysis/scripts/studycourse_analysis.py` — Python 分析脚本 ### 四步分析 1. **基础信息**：角色姓名/年龄/账号ID/手机号后4位/注册时间/购买渠道/设备/首末次完课 2. **完课耗时**：平均值/中位数、异常检测(<10min / >20min)、前后半段趋势 3. **中互动正确率**：Perfect/Good/Oops/Pass/Failed 占比和趋势 4. **知识巩固**：完成率、正确率得分分布 ### 数据源 | 类型 | 库 | 表 | 用途 | |------|-----|-----|------| | MySQL vala_user | vala_app_character | 角色信息、pu",
|
||||
"recallCount": 4,
|
||||
"recallCount": 6,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 4,
|
||||
"totalScore": 6,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-25T05:47:41.388Z",
|
||||
"lastRecalledAt": "2026-05-28T09:07:57.953Z",
|
||||
"lastRecalledAt": "2026-05-28T12:30:30.129Z",
|
||||
"queryHashes": [
|
||||
"9aff8ec9594a",
|
||||
"566b5958861e",
|
||||
"71463fe40be2",
|
||||
"c6c7ff4ed75d"
|
||||
"c6c7ff4ed75d",
|
||||
"7e2572c3140a",
|
||||
"4596e377d39b"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-25",
|
||||
@ -279,18 +283,20 @@
|
||||
"endLine": 16,
|
||||
"source": "memory",
|
||||
"snippet": "# 2026-05-13 工作日志 ## 文档权限规则修正 [Cris 确认] **问题：** AGENTS.md 中「权限告知规则」未区分用户身份，对所有用户（包括业务负责人李应瑛）一视同仁地回复「请添加Bot为知识空间成员」。这导致业务负责人被不必要的技术细节阻塞。 **修正：** 更新 AGENTS.md 第3条权限告知规则，按用户身份分级处理： - 业务负责人（刘庆逊、李应瑛）→ 不告知权限问题，直接联系 Cris 处理 - 其他用户 → 保持原有提示 **验证：** 对 `小斑` 文档（Tn23wQkUQilduAkvgwscTGhgnUd）执行了完整的读/写（追加）/删除测试，Bot身份权限全部正常。 ## lark-cli 绑定 完成了 lark-cli 与 OpenClaw 的绑定（bot-only 模式），后续飞书 API 调用无需额外配置。",
|
||||
"recallCount": 1,
|
||||
"recallCount": 2,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 1,
|
||||
"totalScore": 2,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T07:37:09.223Z",
|
||||
"lastRecalledAt": "2026-05-28T07:37:09.223Z",
|
||||
"lastRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"queryHashes": [
|
||||
"7031af54381b"
|
||||
"7031af54381b",
|
||||
"f22544a8757c"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-28"
|
||||
"2026-05-28",
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"agents.md",
|
||||
@ -364,6 +370,449 @@
|
||||
"unmerge-cells",
|
||||
"update-dimension"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:147:179": {
|
||||
"key": "memory:memory/2026-05-28.md:147:179",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 147,
|
||||
"endLine": 179,
|
||||
"source": "memory",
|
||||
"snippet": "|---|------|------|----------| | 1a | 微伴 → 主表 C 列 | ✅ | 89 行一致 | | 1a2 | 微伴 → 线索分配 2aNzzy | ✅ | 440 行一致 | | 1x | 小溪行课 | ✅ | ⚠ 需Bot交互 | | 1b | 聚光 agent → 主表 | ✅ | 12 格一致 | | 1b2 | 聚光在投笔记 | ✅ | 3 格(55/30/38)一致 | | 1c | 聚光日明细 | ✅ | 137 笔记一致 | | 2b | 客户主表 → 订单 | ✅ | 公式就绪 | | 3 | 日报 C1HVN2 | ✅ | 四段刷新 | | 3a/3c | 口径审计 | ✅ | 推群全过 | | 4a | 结算月汇总 | ✅ | 公式就绪 | **结论：当前表内所有数据与重拉完全一致，零差异，无需刷写。** ### 18:36 小溪行课触发实跑 - 已推 1920 条到小溪查询表（新 Sheet） - 已 @小溪 到小红书数据需求群，请处理 ~5 条待查询记录 - 表格：`RFIJsXT8FhGHhctY4RwczcOfnac` - 等待小溪回填后跑 `pull_xiaoxi_results.py` 回收结果 # 2026-05-28 工作日志 ## 品牌更名与定位升级 [Cris确认] - 姓名：小斑 → **大麦** - 定位：AI班主任 → **增长营销分析师** - 核心职能：增长数据分析、营销效果评估、转化漏斗分析、商业化洞察 - Emoji：📚 → 🌾 - 已更新文件：IDENTITY.md、AGE",
|
||||
"recallCount": 3,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 3,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T12:29:25.867Z",
|
||||
"lastRecalledAt": "2026-05-28T12:30:30.129Z",
|
||||
"queryHashes": [
|
||||
"a72168a828c9",
|
||||
"7e2572c3140a",
|
||||
"4596e377d39b"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-28"
|
||||
],
|
||||
"conceptTags": [
|
||||
"55/30/38",
|
||||
"3a/3c",
|
||||
"pull-xiaoxi-results.py",
|
||||
"identity.md",
|
||||
"一致",
|
||||
"1a2",
|
||||
"线索",
|
||||
"分配"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:313:345": {
|
||||
"key": "memory:memory/2026-05-28.md:313:345",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 313,
|
||||
"endLine": 345,
|
||||
"source": "memory",
|
||||
"snippet": "|---|------|------|----------| | 1a | 微伴 → 主表 C 列 | ✅ | 89 行一致 | | 1a2 | 微伴 → 线索分配 2aNzzy | ✅ | 440 行一致 | | 1x | 小溪行课 | ✅ | ⚠ 需Bot交互 | | 1b | 聚光 agent → 主表 | ✅ | 12 格一致 | | 1b2 | 聚光在投笔记 | ✅ | 3 格(55/30/38)一致 | | 1c | 聚光日明细 | ✅ | 137 笔记一致 | | 2b | 客户主表 → 订单 | ✅ | 公式就绪 | | 3 | 日报 C1HVN2 | ✅ | 四段刷新 | | 3a/3c | 口径审计 | ✅ | 推群全过 | | 4a | 结算月汇总 | ✅ | 公式就绪 | **结论：当前表内所有数据与重拉完全一致，零差异，无需刷写。** ### 18:36 小溪行课触发实跑 - 已推 1920 条到小溪查询表（新 Sheet） - 已 @小溪 到小红书数据需求群，请处理 ~5 条待查询记录 - 表格：`RFIJsXT8FhGHhctY4RwczcOfnac` - 等待小溪回填后跑 `pull_xiaoxi_results.py` 回收结果 ### 18:50 PG 实时行课数据全量覆盖 [陈逸鸫确认] 不用等小溪回填那 5 条，直接用 PG `user_course_detail` 表实时数据覆盖： - 448 人有新进展（相比小溪历史快照） - pull → 3PRySY 口径对齐 12/12 ✅ - lesson_cache → C1HVN",
|
||||
"recallCount": 3,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 3,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T12:29:25.867Z",
|
||||
"lastRecalledAt": "2026-05-28T12:30:30.129Z",
|
||||
"queryHashes": [
|
||||
"a72168a828c9",
|
||||
"7e2572c3140a",
|
||||
"4596e377d39b"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-28"
|
||||
],
|
||||
"conceptTags": [
|
||||
"55/30/38",
|
||||
"3a/3c",
|
||||
"pull-xiaoxi-results.py",
|
||||
"user-course-detail",
|
||||
"12/12",
|
||||
"lesson-cache",
|
||||
"一致",
|
||||
"1a2"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:122:152": {
|
||||
"key": "memory:memory/2026-05-28.md:122:152",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 122,
|
||||
"endLine": 152,
|
||||
"source": "memory",
|
||||
"snippet": "### 17:56 大麦查询输出表就绪 ✅ [陈逸鸫] 输出表 Token: `GCqNsqgzKhfQ5atFLnQcmLcGn4d`，Sheet: 查询快照 (`fd42b8`) - 实习虾 `cli_aa898f32d4799bea` 已分享编辑权限 - 写入验证通过 (revision 4) - env var: `XIAOBAN_QUERY_OUTPUT_TOKEN=GCqNsqgzKhfQ5atFLnQcmLcGn4d` - 同事查数结果 append `fd42b8!A:G`（请求时间/请求人/查询类型/cutoff/查询参数/结果摘要/备注） ### 17:48 5/27 日报推群成功 ✅ wrapper v0.4 实现 `im +messages-send`，推送 `message_id=om_x100b6eb54f9da0f4b115725198feede` ### 待办汇总 1. 📋 陈逸鸫发 `docs/xiaoban-data-boundary.md` §7 Agent 系统提示词 2. 📋 陈逸鸫发 `docs/xiaoban-runbook.md` §2 部署 checklist 3. 📋 建「大麦查询输出」飞书表 → 陈逸鸫提供 token 4. 📋 `~/xhs-tech-dashboard` 仓库 clone → 陈逸鸫提供访问方式 5. ⏳ 全量 pipeline 聚光步骤验证（子进程中） ### 18:30 Pipeline 全链路 dry-run 验证全部通过 [陈逸鸫] 收到微伴 xlsx，触发全链路 dry-r",
|
||||
"recallCount": 2,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 2,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T12:30:30.129Z",
|
||||
"lastRecalledAt": "2026-05-28T20:57:42.055Z",
|
||||
"queryHashes": [
|
||||
"4596e377d39b",
|
||||
"f139ebdae100"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-28",
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"cli-aa898f32d4799bea",
|
||||
"xiaoban-query-output-token",
|
||||
"5/27",
|
||||
"v0.4",
|
||||
"messages-send",
|
||||
"message-id",
|
||||
"docs/xiaoban-data-boundary.md",
|
||||
"docs/xiaoban-runbook.md"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:288:318": {
|
||||
"key": "memory:memory/2026-05-28.md:288:318",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 288,
|
||||
"endLine": 318,
|
||||
"source": "memory",
|
||||
"snippet": "### 17:56 大麦查询输出表就绪 ✅ [陈逸鸫] 输出表 Token: `GCqNsqgzKhfQ5atFLnQcmLcGn4d`，Sheet: 查询快照 (`fd42b8`) - 实习虾 `cli_aa898f32d4799bea` 已分享编辑权限 - 写入验证通过 (revision 4) - env var: `XIAOBAN_QUERY_OUTPUT_TOKEN=GCqNsqgzKhfQ5atFLnQcmLcGn4d` - 同事查数结果 append `fd42b8!A:G`（请求时间/请求人/查询类型/cutoff/查询参数/结果摘要/备注） ### 17:48 5/27 日报推群成功 ✅ wrapper v0.4 实现 `im +messages-send`，推送 `message_id=om_x100b6eb54f9da0f4b115725198feede` ### 待办汇总 1. 📋 陈逸鸫发 `docs/xiaoban-data-boundary.md` §7 Agent 系统提示词 2. 📋 陈逸鸫发 `docs/xiaoban-runbook.md` §2 部署 checklist 3. 📋 建「大麦查询输出」飞书表 → 陈逸鸫提供 token 4. 📋 `~/xhs-tech-dashboard` 仓库 clone → 陈逸鸫提供访问方式 5. ⏳ 全量 pipeline 聚光步骤验证（子进程中） ### 18:30 Pipeline 全链路 dry-run 验证全部通过 [陈逸鸫] 收到微伴 xlsx，触发全链路 dry-r",
|
||||
"recallCount": 1,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 1,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T12:30:30.129Z",
|
||||
"lastRecalledAt": "2026-05-28T12:30:30.129Z",
|
||||
"queryHashes": [
|
||||
"4596e377d39b"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-28"
|
||||
],
|
||||
"conceptTags": [
|
||||
"cli-aa898f32d4799bea",
|
||||
"xiaoban-query-output-token",
|
||||
"5/27",
|
||||
"v0.4",
|
||||
"messages-send",
|
||||
"message-id",
|
||||
"docs/xiaoban-data-boundary.md",
|
||||
"docs/xiaoban-runbook.md"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:504:536": {
|
||||
"key": "memory:memory/2026-05-28.md:504:536",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 504,
|
||||
"endLine": 536,
|
||||
"source": "memory",
|
||||
"snippet": "| 2b | 客户主表 → 订单 | ✅ | 公式就绪 | | 3 | 日报 C1HVN2 | ✅ | 四段刷新 | | 3a/3c | 口径审计 | ✅ | 推群全过 | | 4a | 结算月汇总 | ✅ | 公式就绪 | **结论：当前表内所有数据与重拉完全一致，零差异，无需刷写。** ### 18:36 小溪行课触发实跑 - 已推 1920 条到小溪查询表（新 Sheet） - 已 @小溪 到小红书数据需求群，请处理 ~5 条待查询记录 - 表格：`RFIJsXT8FhGHhctY4RwczcOfnac` - 等待小溪回填后跑 `pull_xiaoxi_results.py` 回收结果 ### 18:50 PG 实时行课数据全量覆盖 [陈逸鸫确认] 不用等小溪回填那 5 条，直接用 PG `user_course_detail` 表实时数据覆盖： - 448 人有新进展（相比小溪历史快照） - pull → 3PRySY 口径对齐 12/12 ✅ - lesson_cache → C1HVN2 16 格 ✅ - 变化有限（PG=正式课，小溪=体验课，口径不同） - sync_base (4b) → 多维表格 8 张 ✅ 21.2s ### 18:54 漏斗看板发布待解决 - funnel HTML 已构建（scripts/build_funnel_dashboard.py ✅） - 妙搭发布 `apps +html-publish` 需要 `--as user`，bot 模式不支持 - 系统 lark-cli `/usr/local/lib/node_mod",
|
||||
"recallCount": 1,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 1,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"lastRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"queryHashes": [
|
||||
"f22544a8757c"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"3a/3c",
|
||||
"pull-xiaoxi-results.py",
|
||||
"user-course-detail",
|
||||
"12/12",
|
||||
"lesson-cache",
|
||||
"sync-base",
|
||||
"21.2s",
|
||||
"html-publish"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:337:366": {
|
||||
"key": "memory:memory/2026-05-28.md:337:366",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 337,
|
||||
"endLine": 366,
|
||||
"source": "memory",
|
||||
"snippet": "- 448 人有新进展（相比小溪历史快照） - pull → 3PRySY 口径对齐 12/12 ✅ - lesson_cache → C1HVN2 16 格 ✅ - 变化有限（PG=正式课，小溪=体验课，口径不同） - sync_base (4b) → 多维表格 8 张 ✅ 21.2s ### 18:54 漏斗看板发布待解决 - funnel HTML 已构建（scripts/build_funnel_dashboard.py ✅） - 妙搭发布 `apps +html-publish` 需要 `--as user`，bot 模式不支持 - 系统 lark-cli `/usr/local/lib/node_modules/@anthropic/lark-cli` 可能支持 - 待确认：服务器是否已 `lark-cli auth login --as user` ### 19:00 陈逸鸫派 Image2 生图任务 **任务：** L1-S1-U1《秘密基地》5 课投放用小地图底图 - 模型：gpt-image-2 · 3:4 · 2K · 不要文字 - 风格：太阳朋克 + L1 场景 - 5 张 PNG：U1-L1～U1-L5 - FUNCLOUD_API_KEY 在小研 workspace `.env` 中可用（`fc_eea138933b02b4797ce0779ffb637d8b8a6368db7b435dfdab7b4be1cd254d98`） - Brief 文档 `/docx/KsVadUTmooO7yYxHaGmc1R0Bn5b` + 投放手册 `/do",
|
||||
"recallCount": 1,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 1,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"lastRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"queryHashes": [
|
||||
"f22544a8757c"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"gpt",
|
||||
"12/12",
|
||||
"lesson-cache",
|
||||
"sync-base",
|
||||
"21.2s",
|
||||
"html-publish",
|
||||
"lark-cli",
|
||||
"usr/local/lib/node-modules"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:74:100": {
|
||||
"key": "memory:memory/2026-05-28.md:74:100",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 74,
|
||||
"endLine": 100,
|
||||
"source": "memory",
|
||||
"snippet": "3. **check_call→run**（`sync_juguang_notes.py` `sheet_write_range`）：`subprocess.check_call` 不支持 `input=`, 改用 `subprocess.run(..., check=True, input=...)` 1b 单天验证通过：`sync_juguang_agents.py --start 2026-05-27 --end 2026-05-27` 写入 12 个单元格成功。全量 27 天在子进程跑，结果待出。 ### 17:12 数据服务边界规则部署 [陈逸鸫] **文档位置：** `docs/xiaoban-data-boundary.md`（git@github.com:chenyd11/feishu-database.git — 服务器无 SSH key 无法 pull） **三条核心规则（已写入 MEMORY.md）：** 1. cron pipeline 和帮同事查数分轨，不能混用同一流程 2. 同事请求默认只读；写生产表 CYFTsu 必须 @陈逸鸫 确认 3. 查数结果写「输出区」副本，不改主表 **操作黑名单：** `pipeline.py` / `sync_*` / `run_juguang_*` / `sheets +write`(生产表) / `--promote` **待办：** 建「大麦查询输出」专用表（待陈逸鸫提供 token） **待办：** 获取 docs/xiaoban-data-boundary.md §7 Agent 系统提示词（完",
|
||||
"recallCount": 1,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 1,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"lastRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"queryHashes": [
|
||||
"f22544a8757c"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"check-call",
|
||||
"sync-juguang-notes.py",
|
||||
"sheet-write-range",
|
||||
"subprocess.check-call",
|
||||
"subprocess.run",
|
||||
"sync-juguang-agents.py",
|
||||
"docs/xiaoban-data-boundary.md",
|
||||
"github.com"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:240:266": {
|
||||
"key": "memory:memory/2026-05-28.md:240:266",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 240,
|
||||
"endLine": 266,
|
||||
"source": "memory",
|
||||
"snippet": "3. **check_call→run**（`sync_juguang_notes.py` `sheet_write_range`）：`subprocess.check_call` 不支持 `input=`, 改用 `subprocess.run(..., check=True, input=...)` 1b 单天验证通过：`sync_juguang_agents.py --start 2026-05-27 --end 2026-05-27` 写入 12 个单元格成功。全量 27 天在子进程跑，结果待出。 ### 17:12 数据服务边界规则部署 [陈逸鸫] **文档位置：** `docs/xiaoban-data-boundary.md`（git@github.com:chenyd11/feishu-database.git — 服务器无 SSH key 无法 pull） **三条核心规则（已写入 MEMORY.md）：** 1. cron pipeline 和帮同事查数分轨，不能混用同一流程 2. 同事请求默认只读；写生产表 CYFTsu 必须 @陈逸鸫 确认 3. 查数结果写「输出区」副本，不改主表 **操作黑名单：** `pipeline.py` / `sync_*` / `run_juguang_*` / `sheets +write`(生产表) / `--promote` **待办：** 建「大麦查询输出」专用表（待陈逸鸫提供 token） **待办：** 获取 docs/xiaoban-data-boundary.md §7 Agent 系统提示词（完",
|
||||
"recallCount": 1,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 1,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"lastRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"queryHashes": [
|
||||
"f22544a8757c"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"check-call",
|
||||
"sync-juguang-notes.py",
|
||||
"sheet-write-range",
|
||||
"subprocess.check-call",
|
||||
"subprocess.run",
|
||||
"sync-juguang-agents.py",
|
||||
"docs/xiaoban-data-boundary.md",
|
||||
"github.com"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:432:458": {
|
||||
"key": "memory:memory/2026-05-28.md:432:458",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 432,
|
||||
"endLine": 458,
|
||||
"source": "memory",
|
||||
"snippet": "3. **check_call→run**（`sync_juguang_notes.py` `sheet_write_range`）：`subprocess.check_call` 不支持 `input=`, 改用 `subprocess.run(..., check=True, input=...)` 1b 单天验证通过：`sync_juguang_agents.py --start 2026-05-27 --end 2026-05-27` 写入 12 个单元格成功。全量 27 天在子进程跑，结果待出。 ### 17:12 数据服务边界规则部署 [陈逸鸫] **文档位置：** `docs/xiaoban-data-boundary.md`（git@github.com:chenyd11/feishu-database.git — 服务器无 SSH key 无法 pull） **三条核心规则（已写入 MEMORY.md）：** 1. cron pipeline 和帮同事查数分轨，不能混用同一流程 2. 同事请求默认只读；写生产表 CYFTsu 必须 @陈逸鸫 确认 3. 查数结果写「输出区」副本，不改主表 **操作黑名单：** `pipeline.py` / `sync_*` / `run_juguang_*` / `sheets +write`(生产表) / `--promote` **待办：** 建「大麦查询输出」专用表（待陈逸鸫提供 token） **待办：** 获取 docs/xiaoban-data-boundary.md §7 Agent 系统提示词（完",
|
||||
"recallCount": 1,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 1,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"lastRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"queryHashes": [
|
||||
"f22544a8757c"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"check-call",
|
||||
"sync-juguang-notes.py",
|
||||
"sheet-write-range",
|
||||
"subprocess.check-call",
|
||||
"subprocess.run",
|
||||
"sync-juguang-agents.py",
|
||||
"docs/xiaoban-data-boundary.md",
|
||||
"github.com"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:530:557": {
|
||||
"key": "memory:memory/2026-05-28.md:530:557",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 530,
|
||||
"endLine": 557,
|
||||
"source": "memory",
|
||||
"snippet": "- 妙搭发布 `apps +html-publish` 需要 `--as user`，bot 模式不支持 - 系统 lark-cli `/usr/local/lib/node_modules/@anthropic/lark-cli` 可能支持 - 待确认：服务器是否已 `lark-cli auth login --as user` ### 19:00 陈逸鸫派 Image2 生图任务 **任务：** L1-S1-U1《秘密基地》5 课投放用小地图底图 - 模型：gpt-image-2 · 3:4 · 2K · 不要文字 - 风格：太阳朋克 + L1 场景 - 5 张 PNG：U1-L1～U1-L5 - FUNCLOUD_API_KEY 在小研 workspace `.env` 中可用（`fc_eea138933b02b4797ce0779ffb637d8b8a6368db7b435dfdab7b4be1cd254d98`） - Brief 文档 `/docx/KsVadUTmooO7yYxHaGmc1R0Bn5b` + 投放手册 `/docx/QhYQdz7PvoN7Eaxmhu0c0Q5UnHe` — 均为个人文档，AGENTS.md 规则禁止读取 - 素材库入口：https://llm-dev.valavala.com/web_tools/material_prod --- ### 19:30 同事数据查询流程演练 [陈逸鸫测试] **场景：模拟王虹茗请求小龙 4/21-5/20 订单详情，验证三级查询流程** **小龙订单查询结果（数据源：3wcle8 销售订",
|
||||
"recallCount": 1,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 1,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"lastRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"queryHashes": [
|
||||
"f22544a8757c"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"gpt",
|
||||
"html-publish",
|
||||
"lark-cli",
|
||||
"usr/local/lib/node-modules",
|
||||
"anthropic/lark-cli",
|
||||
"l1-s1-u1",
|
||||
"gpt-image-2",
|
||||
"u1-l1"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:94:127": {
|
||||
"key": "memory:memory/2026-05-28.md:94:127",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 94,
|
||||
"endLine": 127,
|
||||
"source": "memory",
|
||||
"snippet": "**文档：** `docs/xiaoban-runbook.md` §2 | 时间 | 任务 | 状态 | |------|------|------| | 07:15 | deploy_funnel_dashboard (Base + 销售漏斗) | ❓ 需部署 | | 09:50 | deploy_cockpit (Base 二次 + 投放指挥舱) | ⚠️ 需 `~/xhs-tech-dashboard` 仓库 | | 5a | 聚光实时 | ❌ 未进 cron | | 5b/5e | 销售看板 | ❌ 未进 cron | `xhs-tech-dashboard` 仓库需要拉到服务器（路径 `~/xhs-tech-dashboard`），待陈逸鸫提供。 ### 17:36 cron 已挂 ✅ [陈逸鸫确认] 三条工作日 crontab： ``` 15 7 * * 1-5 run_xiaoxi_daily.sh # 小溪+Base+漏斗早刷 5 9 * * 1-5 run_juguang_core.sh # 1b+1c+3+审计→DM日报 50 9 * * 1-5 run_juguang_slow.sh # 1b2+Base二次+全量看板 ``` - 微伴段：非 cron，收到 DM xlsx 后触发 `run_morning_weiban_from_dm.sh` - 推群：人工，不自动 `push_daily_report_to_group` - Git remote 已配 token URL：`https://gho_***",
|
||||
"recallCount": 2,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 2,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"lastRecalledAt": "2026-05-28T20:57:42.055Z",
|
||||
"queryHashes": [
|
||||
"f22544a8757c",
|
||||
"f139ebdae100"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"docs/xiaoban-runbook.md",
|
||||
"deploy-funnel-dashboard",
|
||||
"deploy-cockpit",
|
||||
"xhs-tech-dashboard",
|
||||
"5b/5e",
|
||||
"1-5",
|
||||
"run-xiaoxi-daily.sh",
|
||||
"run-juguang-core.sh"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:260:293": {
|
||||
"key": "memory:memory/2026-05-28.md:260:293",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 260,
|
||||
"endLine": 293,
|
||||
"source": "memory",
|
||||
"snippet": "**文档：** `docs/xiaoban-runbook.md` §2 | 时间 | 任务 | 状态 | |------|------|------| | 07:15 | deploy_funnel_dashboard (Base + 销售漏斗) | ❓ 需部署 | | 09:50 | deploy_cockpit (Base 二次 + 投放指挥舱) | ⚠️ 需 `~/xhs-tech-dashboard` 仓库 | | 5a | 聚光实时 | ❌ 未进 cron | | 5b/5e | 销售看板 | ❌ 未进 cron | `xhs-tech-dashboard` 仓库需要拉到服务器（路径 `~/xhs-tech-dashboard`），待陈逸鸫提供。 ### 17:36 cron 已挂 ✅ [陈逸鸫确认] 三条工作日 crontab： ``` 15 7 * * 1-5 run_xiaoxi_daily.sh # 小溪+Base+漏斗早刷 5 9 * * 1-5 run_juguang_core.sh # 1b+1c+3+审计→DM日报 50 9 * * 1-5 run_juguang_slow.sh # 1b2+Base二次+全量看板 ``` - 微伴段：非 cron，收到 DM xlsx 后触发 `run_morning_weiban_from_dm.sh` - 推群：人工，不自动 `push_daily_report_to_group` - Git remote 已配 token URL：`https://gho_***",
|
||||
"recallCount": 2,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 2,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"lastRecalledAt": "2026-05-28T20:57:42.055Z",
|
||||
"queryHashes": [
|
||||
"f22544a8757c",
|
||||
"f139ebdae100"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"docs/xiaoban-runbook.md",
|
||||
"deploy-funnel-dashboard",
|
||||
"deploy-cockpit",
|
||||
"xhs-tech-dashboard",
|
||||
"5b/5e",
|
||||
"1-5",
|
||||
"run-xiaoxi-daily.sh",
|
||||
"run-juguang-core.sh"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-28.md:452:485": {
|
||||
"key": "memory:memory/2026-05-28.md:452:485",
|
||||
"path": "memory/2026-05-28.md",
|
||||
"startLine": 452,
|
||||
"endLine": 485,
|
||||
"source": "memory",
|
||||
"snippet": "**文档：** `docs/xiaoban-runbook.md` §2 | 时间 | 任务 | 状态 | |------|------|------| | 07:15 | deploy_funnel_dashboard (Base + 销售漏斗) | ❓ 需部署 | | 09:50 | deploy_cockpit (Base 二次 + 投放指挥舱) | ⚠️ 需 `~/xhs-tech-dashboard` 仓库 | | 5a | 聚光实时 | ❌ 未进 cron | | 5b/5e | 销售看板 | ❌ 未进 cron | `xhs-tech-dashboard` 仓库需要拉到服务器（路径 `~/xhs-tech-dashboard`），待陈逸鸫提供。 ### 17:36 cron 已挂 ✅ [陈逸鸫确认] 三条工作日 crontab： ``` 15 7 * * 1-5 run_xiaoxi_daily.sh # 小溪+Base+漏斗早刷 5 9 * * 1-5 run_juguang_core.sh # 1b+1c+3+审计→DM日报 50 9 * * 1-5 run_juguang_slow.sh # 1b2+Base二次+全量看板 ``` - 微伴段：非 cron，收到 DM xlsx 后触发 `run_morning_weiban_from_dm.sh` - 推群：人工，不自动 `push_daily_report_to_group` - Git remote 已配 token URL：`https://gho_***",
|
||||
"recallCount": 2,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 2,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:51:03.908Z",
|
||||
"lastRecalledAt": "2026-05-28T20:57:42.055Z",
|
||||
"queryHashes": [
|
||||
"f22544a8757c",
|
||||
"f139ebdae100"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"docs/xiaoban-runbook.md",
|
||||
"deploy-funnel-dashboard",
|
||||
"deploy-cockpit",
|
||||
"xhs-tech-dashboard",
|
||||
"5b/5e",
|
||||
"1-5",
|
||||
"run-xiaoxi-daily.sh",
|
||||
"run-juguang-core.sh"
|
||||
]
|
||||
},
|
||||
"memory:memory/2026-05-24.md:122:132": {
|
||||
"key": "memory:memory/2026-05-24.md:122:132",
|
||||
"path": "memory/2026-05-24.md",
|
||||
"startLine": 122,
|
||||
"endLine": 132,
|
||||
"source": "memory",
|
||||
"snippet": "- HTML 模板使用 f-string 时，CSS 中的 `{` 必须双写 `{{`，且不能在 f-string 内做字符串拼接（会打断 f-string） - 年龄从 birthday（varchar \"YYYY-MM-DD\"）计算，非 DATE 类型 - 手机号在 MySQL 已脱敏（如 186****1625），直接取后4位 ### 已测试角色 - **32009 (zyl)**：26 条完课，651 条中互动(Perfect 82.6%、Good 2.9%)，26 节巩固全部满分。课程 A2，Apple App Store 购买，渠道 newmedia-daren-xhs-宣儿麻麻。设备小课屏E3。学习时间 2026-04-24 ~ 2026-05-24。 ### 同步 - 已推送到 SkillHub（`studycourse-analysis.xiaoban`） - 已 commit 8648b7b 到 Git 远程仓库 - 已通知 Cris（李若松）",
|
||||
"recallCount": 1,
|
||||
"dailyCount": 0,
|
||||
"groundedCount": 0,
|
||||
"totalScore": 1,
|
||||
"maxScore": 1,
|
||||
"firstRecalledAt": "2026-05-28T20:57:42.055Z",
|
||||
"lastRecalledAt": "2026-05-28T20:57:42.055Z",
|
||||
"queryHashes": [
|
||||
"f139ebdae100"
|
||||
],
|
||||
"recallDays": [
|
||||
"2026-05-29"
|
||||
],
|
||||
"conceptTags": [
|
||||
"f-string",
|
||||
"yyyy-mm-dd",
|
||||
"82.6",
|
||||
"2.9",
|
||||
"newmedia-daren-xhs-宣儿麻麻",
|
||||
"studycourse-analysis.xiaoban",
|
||||
"html",
|
||||
"模板"
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
144
memory/2026-05-28-2348.md
Normal file
144
memory/2026-05-28-2348.md
Normal file
@ -0,0 +1,144 @@
|
||||
# Session: 2026-05-28 23:48:37 GMT+8
|
||||
|
||||
- **Session Key**: agent:xiaoban:feishu:direct:ou_f981d4811369c954b3597908ca93a01c
|
||||
- **Session ID**: f435db07-4b5f-4315-a87d-76d6b0d06650
|
||||
- **Source**: feishu
|
||||
|
||||
## Conversation Summary
|
||||
|
||||
user: 高潜力怎么会没有呢，你再搜索一下那个表格，肯定有，是按用户的过往课史判断的。手机号脱敏吧，群就是你发日报的那个，一样的
|
||||
assistant: 找到了！
|
||||
|
||||
- **O 列 (idx 14)**: `用户意向(自动)` → "体验4-5节→高意向"
|
||||
- **R 列 (idx 17)**: `用户潜力(自动)` → "课史/跟进含KET→高潜力"
|
||||
|
||||
高潜力是根据 G 列（过往课史）和 H 列（跟进记录）中是否包含"KET"关键词来判定的。让我看看分布：
|
||||
assistant: 完整数据：
|
||||
|
||||
| 销售 | 高意向 | 高潜力 |
|
||||
|------|--------|--------|
|
||||
| 小龙 | 54 | 30 |
|
||||
| 吴迪 | 50 | 30 |
|
||||
| Bob | 12 | — |
|
||||
| Tom | 18 | — |
|
||||
|
||||
**判定逻辑确认：**
|
||||
- **O 列 高意向**：体验4-5节（行课深度高）
|
||||
- **R 列 高潜力**：过往课史/跟进记录包含"KET"关键词（目标明确、有备考需求）
|
||||
|
||||
成都 sheet（Bob/Tom）没有高潜力标记——可能是公式没覆盖到，或者他们的客户确实没有 KET 相关课史。
|
||||
|
||||
现在整个方案清楚了：
|
||||
- 数据源：`ERCFsFo4MhnF0ytGeCrc0Bb8n5f` 三个子 sheet
|
||||
- 字段：D(昵称)、E(手机号)、I(用户ID)、N(行课状态)、O(高意向)、R(高潜力)
|
||||
- 推送目标：日报推群同一个群（销转日会群）
|
||||
- 触发：日报确认后一起推
|
||||
|
||||
让我确认推群的 chat_id：
|
||||
assistant: Good. 销转日会群 = `oc_67d83d28be454517616c6c08699c19ba`
|
||||
|
||||
现在所有信息齐全，让我回复确认。找到了！R 列（index 17）= `用户潜力(自动)`，判定逻辑是"**课史/跟进记录包含 KET → 高潜力**"。
|
||||
|
||||
**当前数据：**
|
||||
|
||||
| 销售 | 高意向（体验4-5节） | 高潜力（课史含KET） |
|
||||
|------|-------------------|-------------------|
|
||||
| 小龙 | 54 | 30 |
|
||||
| 吴迪 | 50 | 30 |
|
||||
| Bob | 12 | 0 |
|
||||
| Tom | 18 | 0 |
|
||||
|
||||
推群目标确认：`oc_67d83d28be454517616c6c08699c19ba`（销转日会群），手机号脱敏。
|
||||
|
||||
我现在去做两件事：
|
||||
1. 29UBVe 格式统一
|
||||
2. 写 `push_high_intent_leads.py` 脚本，挂到 `push_daily_report_to_group.sh` 日报之后自动推
|
||||
user: 去吧，再确认一下明天早上的全部流程，谁做什么
|
||||
assistant: 现在我看到完整的 crontab。明天早上流程如下：
|
||||
|
||||
---
|
||||
|
||||
好的，明天早上（5月29日，周四）完整流程：
|
||||
|
||||
**⏰ 7:15 — `run_xiaoxi_daily.sh`（自动 cron）**
|
||||
| 步骤 | 谁做 | 内容 |
|
||||
|------|------|------|
|
||||
| 1/4 | 🤖 自动 | push 小溪行课请求 + @小溪 |
|
||||
| 2/4 | 🤖 自动 | 等小溪回填 → pull 行课数据 → 日报 §四 |
|
||||
| 3/4 | 🤖 自动 | `compute_lesson_activation.py` → 2aNzzy W/X 列 |
|
||||
| 4/4 | 🤖 自动 | Base 同步 + 销售漏斗看板 build |
|
||||
|
||||
**⏰ 9:30 — `run_juguang_core.sh`（自动 cron）**
|
||||
| 步骤 | 谁做 | 内容 |
|
||||
|------|------|------|
|
||||
| 1 | 🤖 自动 | 聚光 1b+1c（代理商+笔记明细）|
|
||||
| 2 | 🤖 自动 | `compute_lesson_activation.py`（再算一次确保最新）|
|
||||
| 3 | 🤖 自动 | `sync_daily_report.py` → C1HVN2 全部公式刷新 |
|
||||
| 4 | 🤖 自动 | 审计 3a/3c/3d/4c |
|
||||
| 5 | 🤖 自动 | 日报私信给你（陈逸鸫） |
|
||||
|
||||
**⏰ 9:50 — `run_juguang_slow.sh`（自动 cron）**
|
||||
| 步骤 | 谁做 | 内容 |
|
||||
|------|------|------|
|
||||
| | 🤖 自动 | 1b2 + Base 二次 + 全量看板 |
|
||||
|
||||
**⏰ 你确认日报 OK 后 — 手动触发**
|
||||
| 步骤 | 谁做 | 内容 |
|
||||
|------|------|------|
|
||||
| 1 | 👤 你 | 看私信日报，确认没问题 |
|
||||
| 2 | 👤 你 | 跟我说"推群" / 或手动跑 `./scripts/push_daily_report_to_group.sh` |
|
||||
| 3 | 🤖 | 日报推销转日会群 |
|
||||
| 4 | 🤖 | **高意向/高潜力线索推送**（紧跟日报之后）← 新增 |
|
||||
|
||||
---
|
||||
|
||||
**⚠️ 明天有一个已知问题：**
|
||||
- Base 写入被 91403 权限阻挡（系统 lark-cli 绑定坏了），7:15 的 Base 同步可能失败。如果你今晚或明早能在服务器上跑一下 `lark-cli config bind --identity user-default --app-id cli_a92fc074fb5edcb5` 就能修好。不然 Base 暂时不更新（C1HVN2 日报本身不受影响）。
|
||||
|
||||
我现在去写 `push_high_intent_leads.py` + 统一 29UBVe 格式。
|
||||
user: 你把模型切换成gpt 5.5我试试
|
||||
assistant: 目前没有 GPT 5.5 这个模型可用。当前可以切换的模型别名有：
|
||||
|
||||
| 别名 | 模型 |
|
||||
|------|------|
|
||||
| `opus` | Claude Opus 4.6（当前） |
|
||||
| `sonnet` | Claude Sonnet 4.6 |
|
||||
| `gpt5.4` | GPT-5.4 |
|
||||
| `deepseekv4` | DeepSeek V4 Pro |
|
||||
| `gemini` | Gemini 3.1 Pro |
|
||||
| `doubao` | 豆包 Seed 2.0 Pro |
|
||||
|
||||
要切哪个？
|
||||
user: 那就gpt 5.4吧，然后整个流程里，我给你导入微伴呢？
|
||||
assistant: 已切到 GPT-5.4。
|
||||
|
||||
微伴导入的流程——你是每天早上手动把微伴 xlsx 发给我对吧？在明天的流程里：
|
||||
|
||||
**⏰ ~7:00 — 你发微伴 xlsx（手动）**
|
||||
| 步骤 | 谁做 | 内容 |
|
||||
|------|------|------|
|
||||
| 1 | 👤 你 | 从微伴后台导出 xlsx，发到飞书 DM 给我 |
|
||||
| 2 | 🤖 我 | 收到后触发 `run_morning_weiban_from_dm.sh` |
|
||||
| 3 | 🤖 | → 1a（微伴 → 主表 C 列企微总线索） |
|
||||
| 4 | 🤖 | → 1a2（微伴 → 2aNzzy 每日线索分配 + Bob/Tom 分线） |
|
||||
|
||||
这步必须在 7:15 的 `run_xiaoxi_daily.sh` 之前完成，因为小溪查行课依赖 2aNzzy 里的最新线索数据。
|
||||
|
||||
所以完整时间线修正为：
|
||||
|
||||
```
|
||||
~7:00 👤 你发微伴 xlsx
|
||||
🤖 触发 1a + 1a2（微伴→主表→2aNzzy）
|
||||
7:15 🤖 cron: 小溪 push → 等回填 → pull → 行课激活 → Base
|
||||
9:30 🤖 cron: 聚光 → 日报 → 审计 → 私信给你
|
||||
9:50 🤖 cron: 慢线（1b2 + 全量看板）
|
||||
~10:00 👤 你确认日报 → 跟我说"推群"
|
||||
🤖 日报推群 + 高意向线索推群
|
||||
```
|
||||
|
||||
如果你明早来不及 7:00 前发 xlsx 也没关系——`run_xiaoxi_daily.sh` 会用 2aNzzy 表内已有的数据跑，只是当天新进线索不会出现在行课查询里。日报本身不受影响（聚光 + 销转用的是主表/订单明细数据）。
|
||||
user: 7点太早，大概率我是7点多，不会晚于8点，我导入你触发对应流程就行，其他的你照常跑
|
||||
assistant: [assistant turn failed before producing content]
|
||||
user: 7点太早，大概率我是7点多，不会晚于8点，我导入你触发对应流程就行，其他的你照常跑
|
||||
assistant: [assistant turn failed before producing content]
|
||||
user: 模型切换到deepseek
|
||||
9
memory/2026-05-28-2349.md
Normal file
9
memory/2026-05-28-2349.md
Normal file
@ -0,0 +1,9 @@
|
||||
# Session: 2026-05-28 23:49:15 GMT+8
|
||||
|
||||
- **Session Key**: agent:xiaoban:feishu:direct:ou_f981d4811369c954b3597908ca93a01c
|
||||
- **Session ID**: 05cedb04-c7f8-42c6-b943-57e6de4e5068
|
||||
- **Source**: feishu
|
||||
|
||||
## Conversation Summary
|
||||
|
||||
user: 你看下上面的记录，回顾一下我们在干嘛
|
||||
@ -164,3 +164,553 @@ wrapper v0.4 实现 `im +messages-send`，推送 `message_id=om_x100b6eb54f9da0f
|
||||
- 已 @小溪 到小红书数据需求群，请处理 ~5 条待查询记录
|
||||
- 表格：`RFIJsXT8FhGHhctY4RwczcOfnac`
|
||||
- 等待小溪回填后跑 `pull_xiaoxi_results.py` 回收结果
|
||||
# 2026-05-28 工作日志
|
||||
|
||||
## 品牌更名与定位升级 [Cris确认]
|
||||
|
||||
- 姓名：小斑 → **大麦**
|
||||
- 定位：AI班主任 → **增长营销分析师**
|
||||
- 核心职能：增长数据分析、营销效果评估、转化漏斗分析、商业化洞察
|
||||
- Emoji：📚 → 🌾
|
||||
- 已更新文件：IDENTITY.md、AGENTS.md（@规则同步变更）、MEMORY.md
|
||||
- SOUL.md 无需改动（行为方法论为通用框架）
|
||||
|
||||
### 16:20 pipeline 非聚光部分验证完成
|
||||
|
||||
**lark-cli wrapper v0.3 支持的 action：**
|
||||
| 类别 | action | 状态 |
|
||||
|------|--------|------|
|
||||
| sheets | +read, +write, +append, +info, +meta | ✅ |
|
||||
| sheets | +create-sheet, +delete-sheet | ✅ |
|
||||
| sheets | +batch-set-style (stub) | ✅ |
|
||||
| sheets | +merge-cells, +unmerge-cells | ✅ |
|
||||
| sheets | +update-dimension | ✅ |
|
||||
| bitable | +app, +tables, +records, +create, +update | ✅ |
|
||||
| auth | status | ✅ |
|
||||
| im | +messages-send (stub) | ✅ |
|
||||
|
||||
**pipeline 试跑结果（--dry-run）：**
|
||||
| 步骤 | 说明 | 耗时 | 状态 |
|
||||
|------|------|------|------|
|
||||
| 1a/1a2 | 微伴 xlsx | — | ⏭ 跳过（无 xlsx） |
|
||||
| 1b/1b2/1c | 聚光 | — | ⏭ 跳过（无 .env.juguang） |
|
||||
| 2b | 客户主表→订单明细 | 5.1s | ✅ dry-run（真跑数据量大，预计几分钟） |
|
||||
| 3 | 日报 C1HVN2 | 38.3s | ✅ 真写成功 |
|
||||
| 4a | 结算月汇总 | <1s | ✅ 真写成功（零警告） |
|
||||
| 3a/3c/3d/4c | 审计 | — | ⏭ dry-run 跳过 |
|
||||
|
||||
**修复的 wrapper 兼容性问题：**
|
||||
- API 响应格式：`data.spreadsheet` → `data.sheets.sheets`（lark-cli 兼容）
|
||||
- 参数名：`--values` 与 `--data` 等价处理
|
||||
- camelCase→snake_case：`sheetId` → `sheet_id`，`grid_properties` 包装
|
||||
- `+info` 端点：v3 metadata → v2 metainfo（含 sheets 列表）
|
||||
- `update-dimension`：`PUT /dimension_range`，1-indexed
|
||||
|
||||
### 16:48 聚光凭证部署
|
||||
|
||||
**来源：陈逸鸫** — 从 Mac 发来 `.env.juguang.9180` 和 `.env.juguang.9181`，走方案 A（复制文件，不在服务器重 OAuth）
|
||||
|
||||
- 9180（云智 adv=9746532）：refresh 验证通过 ✅
|
||||
- 9181（谦禾 adv=9013261,9598861 / YTL adv=7242040,9891870,10157917,10562529）：refresh 验证通过 ✅
|
||||
- OAuth 回调地址：`https://odourless-demetra-cany.ngrok-free.dev/callback.html`（ngrok 隧道）
|
||||
- `juguang_token.py` 自动 refresh 机制正常
|
||||
|
||||
### 16:56 pipeline 全量试跑 — Linux ARG_MAX 问题
|
||||
|
||||
**命令：** `pipeline.py --cutoff 2026-05-27 --skip 1a,1a2,1x,2c,4b --continue-on-error`
|
||||
|
||||
**失败步骤：**
|
||||
| 步骤 | 失败原因 | 状态 |
|
||||
|------|----------|------|
|
||||
| 1b（聚光代理商） | `sheet_write` rc=1 → 单格范围问题 | ✅ 已修 |
|
||||
| 1b2（聚光活跃笔记） | 同 1b | ✅ 已修 |
|
||||
| 1c（聚光笔记明细） | `OSError: [Errno 7] Argument list too long` — 3146 行 JSON 超 argv 限制 | ✅ 已修 |
|
||||
| 2b（客户主表→订单明细） | rc=1，但 351 行数据实际已写入；可能 unmerge-cells 步骤失败（非致命） | ⚠️ 待验证 |
|
||||
| 3（日报） | — | ✅ 41s |
|
||||
| 3a（审计） | — | ✅ 25s |
|
||||
| 3c（审计） | — | ✅ 29.4s |
|
||||
| 4a（结算月汇总） | — | ✅ 12.9s |
|
||||
| 4c（指标审计） | — | ✅ 7.3s |
|
||||
| 3d | 代理商数据不一致（数据质量问题，非代码） | ❌ |
|
||||
|
||||
**三个底层修复：**
|
||||
|
||||
1. **stdin 大数据传递**（`metrics_contract.py` `sheet_write`）：当 `data_str > 32768` 字节时通过 `input=` 管道传入 subprocess，绕过 Linux ARG_MAX
|
||||
2. **单格范围修复**（`bin/lark-cli-impl`）：`a8d375!M70` → 自动补为 `a8d375!M70:M70`，飞书 sheets API 写操作需要完整范围
|
||||
3. **check_call→run**（`sync_juguang_notes.py` `sheet_write_range`）：`subprocess.check_call` 不支持 `input=`, 改用 `subprocess.run(..., check=True, input=...)`
|
||||
|
||||
1b 单天验证通过：`sync_juguang_agents.py --start 2026-05-27 --end 2026-05-27` 写入 12 个单元格成功。全量 27 天在子进程跑，结果待出。
|
||||
|
||||
### 17:12 数据服务边界规则部署 [陈逸鸫]
|
||||
|
||||
**文档位置：** `docs/xiaoban-data-boundary.md`（git@github.com:chenyd11/feishu-database.git — 服务器无 SSH key 无法 pull）
|
||||
|
||||
**三条核心规则（已写入 MEMORY.md）：**
|
||||
1. cron pipeline 和帮同事查数分轨，不能混用同一流程
|
||||
2. 同事请求默认只读；写生产表 CYFTsu 必须 @陈逸鸫 确认
|
||||
3. 查数结果写「输出区」副本，不改主表
|
||||
|
||||
**操作黑名单：** `pipeline.py` / `sync_*` / `run_juguang_*` / `sheets +write`(生产表) / `--promote`
|
||||
|
||||
**待办：** 建「大麦查询输出」专用表（待陈逸鸫提供 token）
|
||||
**待办：** 获取 docs/xiaoban-data-boundary.md §7 Agent 系统提示词（完整内容）
|
||||
|
||||
### 17:12 Base/看板 cron 分工 [陈逸鸫]
|
||||
|
||||
**文档：** `docs/xiaoban-runbook.md` §2
|
||||
|
||||
| 时间 | 任务 | 状态 |
|
||||
|------|------|------|
|
||||
| 07:15 | deploy_funnel_dashboard (Base + 销售漏斗) | ❓ 需部署 |
|
||||
| 09:50 | deploy_cockpit (Base 二次 + 投放指挥舱) | ⚠️ 需 `~/xhs-tech-dashboard` 仓库 |
|
||||
| 5a | 聚光实时 | ❌ 未进 cron |
|
||||
| 5b/5e | 销售看板 | ❌ 未进 cron |
|
||||
|
||||
`xhs-tech-dashboard` 仓库需要拉到服务器（路径 `~/xhs-tech-dashboard`），待陈逸鸫提供。
|
||||
|
||||
### 17:36 cron 已挂 ✅
|
||||
|
||||
[陈逸鸫确认] 三条工作日 crontab：
|
||||
```
|
||||
15 7 * * 1-5 run_xiaoxi_daily.sh # 小溪+Base+漏斗早刷
|
||||
5 9 * * 1-5 run_juguang_core.sh # 1b+1c+3+审计→DM日报
|
||||
50 9 * * 1-5 run_juguang_slow.sh # 1b2+Base二次+全量看板
|
||||
```
|
||||
- 微伴段：非 cron，收到 DM xlsx 后触发 `run_morning_weiban_from_dm.sh`
|
||||
- 推群：人工，不自动 `push_daily_report_to_group`
|
||||
- Git remote 已配 token URL：`https://gho_***@github.com/chenyd11/feishu-database.git`
|
||||
- 陈逸鸫 Mac launchd 已通知关停
|
||||
|
||||
### 17:34 部署完成验证
|
||||
|
||||
全链路：auth ✅ · juguang 9180/9181 ✅ · sheets read ✅ · deploy_dashboards 脚本就绪 ✅ · xhs-tech-dashboard 解压到 ~/ ✅
|
||||
|
||||
### 17:56 大麦查询输出表就绪 ✅
|
||||
|
||||
[陈逸鸫] 输出表 Token: `GCqNsqgzKhfQ5atFLnQcmLcGn4d`，Sheet: 查询快照 (`fd42b8`)
|
||||
- 实习虾 `cli_aa898f32d4799bea` 已分享编辑权限
|
||||
- 写入验证通过 (revision 4)
|
||||
- env var: `XIAOBAN_QUERY_OUTPUT_TOKEN=GCqNsqgzKhfQ5atFLnQcmLcGn4d`
|
||||
- 同事查数结果 append `fd42b8!A:G`（请求时间/请求人/查询类型/cutoff/查询参数/结果摘要/备注）
|
||||
|
||||
### 17:48 5/27 日报推群成功 ✅
|
||||
|
||||
wrapper v0.4 实现 `im +messages-send`，推送 `message_id=om_x100b6eb54f9da0f4b115725198feede`
|
||||
|
||||
### 待办汇总
|
||||
|
||||
1. 📋 陈逸鸫发 `docs/xiaoban-data-boundary.md` §7 Agent 系统提示词
|
||||
2. 📋 陈逸鸫发 `docs/xiaoban-runbook.md` §2 部署 checklist
|
||||
3. 📋 建「大麦查询输出」飞书表 → 陈逸鸫提供 token
|
||||
4. 📋 `~/xhs-tech-dashboard` 仓库 clone → 陈逸鸫提供访问方式
|
||||
5. ⏳ 全量 pipeline 聚光步骤验证（子进程中）
|
||||
|
||||
### 18:30 Pipeline 全链路 dry-run 验证全部通过
|
||||
|
||||
[陈逸鸫] 收到微伴 xlsx，触发全链路 dry-run 验证：
|
||||
|
||||
| # | 步骤 | 结构 | 数据比对 |
|
||||
|---|------|------|----------|
|
||||
| 1a | 微伴 → 主表 C 列 | ✅ | 89 行一致 |
|
||||
| 1a2 | 微伴 → 线索分配 2aNzzy | ✅ | 440 行一致 |
|
||||
| 1x | 小溪行课 | ✅ | ⚠ 需Bot交互 |
|
||||
| 1b | 聚光 agent → 主表 | ✅ | 12 格一致 |
|
||||
| 1b2 | 聚光在投笔记 | ✅ | 3 格(55/30/38)一致 |
|
||||
| 1c | 聚光日明细 | ✅ | 137 笔记一致 |
|
||||
| 2b | 客户主表 → 订单 | ✅ | 公式就绪 |
|
||||
| 3 | 日报 C1HVN2 | ✅ | 四段刷新 |
|
||||
| 3a/3c | 口径审计 | ✅ | 推群全过 |
|
||||
| 4a | 结算月汇总 | ✅ | 公式就绪 |
|
||||
|
||||
**结论：当前表内所有数据与重拉完全一致，零差异，无需刷写。**
|
||||
|
||||
### 18:36 小溪行课触发实跑
|
||||
|
||||
- 已推 1920 条到小溪查询表（新 Sheet）
|
||||
- 已 @小溪 到小红书数据需求群，请处理 ~5 条待查询记录
|
||||
- 表格：`RFIJsXT8FhGHhctY4RwczcOfnac`
|
||||
- 等待小溪回填后跑 `pull_xiaoxi_results.py` 回收结果
|
||||
|
||||
### 18:50 PG 实时行课数据全量覆盖
|
||||
|
||||
[陈逸鸫确认] 不用等小溪回填那 5 条，直接用 PG `user_course_detail` 表实时数据覆盖：
|
||||
- 448 人有新进展（相比小溪历史快照）
|
||||
- pull → 3PRySY 口径对齐 12/12 ✅
|
||||
- lesson_cache → C1HVN2 16 格 ✅
|
||||
- 变化有限（PG=正式课，小溪=体验课，口径不同）
|
||||
- sync_base (4b) → 多维表格 8 张 ✅ 21.2s
|
||||
|
||||
### 18:54 漏斗看板发布待解决
|
||||
|
||||
- funnel HTML 已构建（scripts/build_funnel_dashboard.py ✅）
|
||||
- 妙搭发布 `apps +html-publish` 需要 `--as user`，bot 模式不支持
|
||||
- 系统 lark-cli `/usr/local/lib/node_modules/@anthropic/lark-cli` 可能支持
|
||||
- 待确认：服务器是否已 `lark-cli auth login --as user`
|
||||
|
||||
### 19:00 陈逸鸫派 Image2 生图任务
|
||||
|
||||
**任务：** L1-S1-U1《秘密基地》5 课投放用小地图底图
|
||||
- 模型：gpt-image-2 · 3:4 · 2K · 不要文字
|
||||
- 风格：太阳朋克 + L1 场景
|
||||
- 5 张 PNG：U1-L1～U1-L5
|
||||
- FUNCLOUD_API_KEY 在小研 workspace `.env` 中可用（`fc_eea138933b02b4797ce0779ffb637d8b8a6368db7b435dfdab7b4be1cd254d98`）
|
||||
- Brief 文档 `/docx/KsVadUTmooO7yYxHaGmc1R0Bn5b` + 投放手册 `/docx/QhYQdz7PvoN7Eaxmhu0c0Q5UnHe` — 均为个人文档，AGENTS.md 规则禁止读取
|
||||
- 素材库入口：https://llm-dev.valavala.com/web_tools/material_prod
|
||||
# 2026-05-28 工作日志
|
||||
|
||||
## 品牌更名与定位升级 [Cris确认]
|
||||
|
||||
- 姓名：小斑 → **大麦**
|
||||
- 定位：AI班主任 → **增长营销分析师**
|
||||
- 核心职能：增长数据分析、营销效果评估、转化漏斗分析、商业化洞察
|
||||
- Emoji：📚 → 🌾
|
||||
- 已更新文件：IDENTITY.md、AGENTS.md（@规则同步变更）、MEMORY.md
|
||||
- SOUL.md 无需改动（行为方法论为通用框架）
|
||||
|
||||
### 16:20 pipeline 非聚光部分验证完成
|
||||
|
||||
**lark-cli wrapper v0.3 支持的 action：**
|
||||
| 类别 | action | 状态 |
|
||||
|------|--------|------|
|
||||
| sheets | +read, +write, +append, +info, +meta | ✅ |
|
||||
| sheets | +create-sheet, +delete-sheet | ✅ |
|
||||
| sheets | +batch-set-style (stub) | ✅ |
|
||||
| sheets | +merge-cells, +unmerge-cells | ✅ |
|
||||
| sheets | +update-dimension | ✅ |
|
||||
| bitable | +app, +tables, +records, +create, +update | ✅ |
|
||||
| auth | status | ✅ |
|
||||
| im | +messages-send (stub) | ✅ |
|
||||
|
||||
**pipeline 试跑结果（--dry-run）：**
|
||||
| 步骤 | 说明 | 耗时 | 状态 |
|
||||
|------|------|------|------|
|
||||
| 1a/1a2 | 微伴 xlsx | — | ⏭ 跳过（无 xlsx） |
|
||||
| 1b/1b2/1c | 聚光 | — | ⏭ 跳过（无 .env.juguang） |
|
||||
| 2b | 客户主表→订单明细 | 5.1s | ✅ dry-run（真跑数据量大，预计几分钟） |
|
||||
| 3 | 日报 C1HVN2 | 38.3s | ✅ 真写成功 |
|
||||
| 4a | 结算月汇总 | <1s | ✅ 真写成功（零警告） |
|
||||
| 3a/3c/3d/4c | 审计 | — | ⏭ dry-run 跳过 |
|
||||
|
||||
**修复的 wrapper 兼容性问题：**
|
||||
- API 响应格式：`data.spreadsheet` → `data.sheets.sheets`（lark-cli 兼容）
|
||||
- 参数名：`--values` 与 `--data` 等价处理
|
||||
- camelCase→snake_case：`sheetId` → `sheet_id`，`grid_properties` 包装
|
||||
- `+info` 端点：v3 metadata → v2 metainfo（含 sheets 列表）
|
||||
- `update-dimension`：`PUT /dimension_range`，1-indexed
|
||||
|
||||
### 16:48 聚光凭证部署
|
||||
|
||||
**来源：陈逸鸫** — 从 Mac 发来 `.env.juguang.9180` 和 `.env.juguang.9181`，走方案 A（复制文件，不在服务器重 OAuth）
|
||||
|
||||
- 9180（云智 adv=9746532）：refresh 验证通过 ✅
|
||||
- 9181（谦禾 adv=9013261,9598861 / YTL adv=7242040,9891870,10157917,10562529）：refresh 验证通过 ✅
|
||||
- OAuth 回调地址：`https://odourless-demetra-cany.ngrok-free.dev/callback.html`（ngrok 隧道）
|
||||
- `juguang_token.py` 自动 refresh 机制正常
|
||||
|
||||
### 16:56 pipeline 全量试跑 — Linux ARG_MAX 问题
|
||||
|
||||
**命令：** `pipeline.py --cutoff 2026-05-27 --skip 1a,1a2,1x,2c,4b --continue-on-error`
|
||||
|
||||
**失败步骤：**
|
||||
| 步骤 | 失败原因 | 状态 |
|
||||
|------|----------|------|
|
||||
| 1b（聚光代理商） | `sheet_write` rc=1 → 单格范围问题 | ✅ 已修 |
|
||||
| 1b2（聚光活跃笔记） | 同 1b | ✅ 已修 |
|
||||
| 1c（聚光笔记明细） | `OSError: [Errno 7] Argument list too long` — 3146 行 JSON 超 argv 限制 | ✅ 已修 |
|
||||
| 2b（客户主表→订单明细） | rc=1，但 351 行数据实际已写入；可能 unmerge-cells 步骤失败（非致命） | ⚠️ 待验证 |
|
||||
| 3（日报） | — | ✅ 41s |
|
||||
| 3a（审计） | — | ✅ 25s |
|
||||
| 3c（审计） | — | ✅ 29.4s |
|
||||
| 4a（结算月汇总） | — | ✅ 12.9s |
|
||||
| 4c（指标审计） | — | ✅ 7.3s |
|
||||
| 3d | 代理商数据不一致（数据质量问题，非代码） | ❌ |
|
||||
|
||||
**三个底层修复：**
|
||||
|
||||
1. **stdin 大数据传递**（`metrics_contract.py` `sheet_write`）：当 `data_str > 32768` 字节时通过 `input=` 管道传入 subprocess，绕过 Linux ARG_MAX
|
||||
2. **单格范围修复**（`bin/lark-cli-impl`）：`a8d375!M70` → 自动补为 `a8d375!M70:M70`，飞书 sheets API 写操作需要完整范围
|
||||
3. **check_call→run**（`sync_juguang_notes.py` `sheet_write_range`）：`subprocess.check_call` 不支持 `input=`, 改用 `subprocess.run(..., check=True, input=...)`
|
||||
|
||||
1b 单天验证通过：`sync_juguang_agents.py --start 2026-05-27 --end 2026-05-27` 写入 12 个单元格成功。全量 27 天在子进程跑，结果待出。
|
||||
|
||||
### 17:12 数据服务边界规则部署 [陈逸鸫]
|
||||
|
||||
**文档位置：** `docs/xiaoban-data-boundary.md`（git@github.com:chenyd11/feishu-database.git — 服务器无 SSH key 无法 pull）
|
||||
|
||||
**三条核心规则（已写入 MEMORY.md）：**
|
||||
1. cron pipeline 和帮同事查数分轨，不能混用同一流程
|
||||
2. 同事请求默认只读；写生产表 CYFTsu 必须 @陈逸鸫 确认
|
||||
3. 查数结果写「输出区」副本，不改主表
|
||||
|
||||
**操作黑名单：** `pipeline.py` / `sync_*` / `run_juguang_*` / `sheets +write`(生产表) / `--promote`
|
||||
|
||||
**待办：** 建「大麦查询输出」专用表（待陈逸鸫提供 token）
|
||||
**待办：** 获取 docs/xiaoban-data-boundary.md §7 Agent 系统提示词（完整内容）
|
||||
|
||||
### 17:12 Base/看板 cron 分工 [陈逸鸫]
|
||||
|
||||
**文档：** `docs/xiaoban-runbook.md` §2
|
||||
|
||||
| 时间 | 任务 | 状态 |
|
||||
|------|------|------|
|
||||
| 07:15 | deploy_funnel_dashboard (Base + 销售漏斗) | ❓ 需部署 |
|
||||
| 09:50 | deploy_cockpit (Base 二次 + 投放指挥舱) | ⚠️ 需 `~/xhs-tech-dashboard` 仓库 |
|
||||
| 5a | 聚光实时 | ❌ 未进 cron |
|
||||
| 5b/5e | 销售看板 | ❌ 未进 cron |
|
||||
|
||||
`xhs-tech-dashboard` 仓库需要拉到服务器（路径 `~/xhs-tech-dashboard`），待陈逸鸫提供。
|
||||
|
||||
### 17:36 cron 已挂 ✅
|
||||
|
||||
[陈逸鸫确认] 三条工作日 crontab：
|
||||
```
|
||||
15 7 * * 1-5 run_xiaoxi_daily.sh # 小溪+Base+漏斗早刷
|
||||
5 9 * * 1-5 run_juguang_core.sh # 1b+1c+3+审计→DM日报
|
||||
50 9 * * 1-5 run_juguang_slow.sh # 1b2+Base二次+全量看板
|
||||
```
|
||||
- 微伴段：非 cron，收到 DM xlsx 后触发 `run_morning_weiban_from_dm.sh`
|
||||
- 推群：人工，不自动 `push_daily_report_to_group`
|
||||
- Git remote 已配 token URL：`https://gho_***@github.com/chenyd11/feishu-database.git`
|
||||
- 陈逸鸫 Mac launchd 已通知关停
|
||||
|
||||
### 17:34 部署完成验证
|
||||
|
||||
全链路：auth ✅ · juguang 9180/9181 ✅ · sheets read ✅ · deploy_dashboards 脚本就绪 ✅ · xhs-tech-dashboard 解压到 ~/ ✅
|
||||
|
||||
### 17:56 大麦查询输出表就绪 ✅
|
||||
|
||||
[陈逸鸫] 输出表 Token: `GCqNsqgzKhfQ5atFLnQcmLcGn4d`，Sheet: 查询快照 (`fd42b8`)
|
||||
- 实习虾 `cli_aa898f32d4799bea` 已分享编辑权限
|
||||
- 写入验证通过 (revision 4)
|
||||
- env var: `XIAOBAN_QUERY_OUTPUT_TOKEN=GCqNsqgzKhfQ5atFLnQcmLcGn4d`
|
||||
- 同事查数结果 append `fd42b8!A:G`（请求时间/请求人/查询类型/cutoff/查询参数/结果摘要/备注）
|
||||
|
||||
### 17:48 5/27 日报推群成功 ✅
|
||||
|
||||
wrapper v0.4 实现 `im +messages-send`，推送 `message_id=om_x100b6eb54f9da0f4b115725198feede`
|
||||
|
||||
### 18:30 Pipeline 全链路 dry-run 验证全部通过
|
||||
|
||||
[陈逸鸫] 收到微伴 xlsx，触发全链路 dry-run 验证：
|
||||
|
||||
| # | 步骤 | 结构 | 数据比对 |
|
||||
|---|------|------|----------|
|
||||
| 1a | 微伴 → 主表 C 列 | ✅ | 89 行一致 |
|
||||
| 1a2 | 微伴 → 线索分配 2aNzzy | ✅ | 440 行一致 |
|
||||
| 1x | 小溪行课 | ✅ | ⚠ 需Bot交互 |
|
||||
| 1b | 聚光 agent → 主表 | ✅ | 12 格一致 |
|
||||
| 1b2 | 聚光在投笔记 | ✅ | 3 格(55/30/38)一致 |
|
||||
| 1c | 聚光日明细 | ✅ | 137 笔记一致 |
|
||||
| 2b | 客户主表 → 订单 | ✅ | 公式就绪 |
|
||||
| 3 | 日报 C1HVN2 | ✅ | 四段刷新 |
|
||||
| 3a/3c | 口径审计 | ✅ | 推群全过 |
|
||||
| 4a | 结算月汇总 | ✅ | 公式就绪 |
|
||||
|
||||
**结论：当前表内所有数据与重拉完全一致，零差异，无需刷写。**
|
||||
|
||||
### 18:36 小溪行课触发实跑
|
||||
|
||||
- 已推 1920 条到小溪查询表（新 Sheet）
|
||||
- 已 @小溪 到小红书数据需求群，请处理 ~5 条待查询记录
|
||||
- 表格：`RFIJsXT8FhGHhctY4RwczcOfnac`
|
||||
- 等待小溪回填后跑 `pull_xiaoxi_results.py` 回收结果
|
||||
|
||||
### 18:50 PG 实时行课数据全量覆盖
|
||||
|
||||
[陈逸鸫确认] 不用等小溪回填那 5 条，直接用 PG `user_course_detail` 表实时数据覆盖：
|
||||
- 448 人有新进展（相比小溪历史快照）
|
||||
- pull → 3PRySY 口径对齐 12/12 ✅
|
||||
- lesson_cache → C1HVN2 16 格 ✅
|
||||
- 变化有限（PG=正式课，小溪=体验课，口径不同）
|
||||
- sync_base (4b) → 多维表格 8 张 ✅ 21.2s
|
||||
|
||||
### 18:54 漏斗看板发布待解决
|
||||
|
||||
- funnel HTML 已构建（scripts/build_funnel_dashboard.py ✅）
|
||||
- 妙搭发布 `apps +html-publish` 需要 `--as user`，bot 模式不支持
|
||||
- 系统 lark-cli `/usr/local/lib/node_modules/@anthropic/lark-cli` 可能支持
|
||||
- 待确认：服务器是否已 `lark-cli auth login --as user`
|
||||
|
||||
### 19:00 陈逸鸫派 Image2 生图任务
|
||||
|
||||
**任务：** L1-S1-U1《秘密基地》5 课投放用小地图底图
|
||||
- 模型：gpt-image-2 · 3:4 · 2K · 不要文字
|
||||
- 风格：太阳朋克 + L1 场景
|
||||
- 5 张 PNG：U1-L1～U1-L5
|
||||
- FUNCLOUD_API_KEY 在小研 workspace `.env` 中可用（`fc_eea138933b02b4797ce0779ffb637d8b8a6368db7b435dfdab7b4be1cd254d98`）
|
||||
- Brief 文档 `/docx/KsVadUTmooO7yYxHaGmc1R0Bn5b` + 投放手册 `/docx/QhYQdz7PvoN7Eaxmhu0c0Q5UnHe` — 均为个人文档，AGENTS.md 规则禁止读取
|
||||
- 素材库入口：https://llm-dev.valavala.com/web_tools/material_prod
|
||||
|
||||
---
|
||||
|
||||
### 19:30 同事数据查询流程演练 [陈逸鸫测试]
|
||||
|
||||
**场景：模拟王虹茗请求小龙 4/21-5/20 订单详情，验证三级查询流程**
|
||||
|
||||
**小龙订单查询结果（数据源：3wcle8 销售订单明细，筛选 "小龙" + 2026/04/21-05/20）：**
|
||||
- 18 单 | GMV ¥44,375 | 退款 ¥0
|
||||
- 渠道：销转 11 / 达人 4 / 端内 3
|
||||
- 客单价：¥599×2 / ¥1,999×9 / ¥3,598×7
|
||||
- 订单日期分布在 11 天（04/23–05/15）
|
||||
|
||||
**看板发布全流程梳理：**
|
||||
- 服务器 build HTML + Base sync → DM 陈逸鸫 → Mac 妙搭 `html-publish` + `access-scope-set`
|
||||
- 三个看板 App ID：漏斗 `app_4k886pmc9x6yt` · 指挥舱 `app_4k79smc6fa1kf` · 销售 `app_4k7qkz9wrga74`
|
||||
- 当前 cron 不自动发布看板，仅 build HTML
|
||||
|
||||
**王虹茗身份确认：** wanghongming@makee.com，user_id 未获取
|
||||
- 尝试 `lark-cli contact +search-user wanghongming@makee.com` → 失败：bot API 缺少 `search:user` scope
|
||||
- 替代方案：让她发消息给大麦（系统自动获取 user_id），或陈逸鸫截图资料页
|
||||
|
||||
**同事查询三级场景定稿：**
|
||||
1. 常规只读 → 直接查，append 输出表
|
||||
2. 权限外用户 → 先通知业务负责人，再决定是否返回
|
||||
3. 写操作 → 回复「这会影响生产数据，已转 @陈逸鸫 确认」
|
||||
|
||||
### 20:20 行课转化分析 [陈逸鸫需求]
|
||||
|
||||
**需求：行课记录新增当日进线→当天行课 + 7天首课率 + 销售排名 + 日报展示**
|
||||
|
||||
**数据摸底结果：**
|
||||
|
||||
| 指标 | 5月数据 | 定义 |
|
||||
|------|---------|------|
|
||||
| 当日进线→当天行课 | 3.0% (4/135) | 进线当天有 chapter settlement |
|
||||
| 7天线索→首课 | 28.6% (10/35) | 过去7天进线中 小溪阶段≥5 |
|
||||
|
||||
**按销售拆解：**
|
||||
| 销售 | 当日行课率 | 7天首课率 |
|
||||
|------|-----------|----------|
|
||||
| Bob | 7.4% (最高) | 28.6% |
|
||||
| Tom | 3.7% | 28.6% |
|
||||
| 吴迪 | 0% | 28.6% |
|
||||
| 小龙 | 0% | 14.3% (最低) |
|
||||
|
||||
**改动方案（6处，待陈逸鸫确认后执行）：**
|
||||
|
||||
1. **2aNzzy 每日线索分配** — 加 2 列（首课日期 V、当日行课 W）
|
||||
2. **3PRySY 行课情况** — 加 4 列（当日行课数、当日行课率、7天线索数、7天首课率）
|
||||
3. **C1HVN2 日报** — 新增「五、行课转化」段
|
||||
4. **Base 多维表格** — 3 张表加字段（tbl9HgnKU2qgj4gE / tblB27fjK7mDnxjY / tblxjCeGOPNoLHl6）
|
||||
5. **funnel-daily 看板 HTML** — 新增「当日转化」+「7天首课」指标
|
||||
6. **build_pipeline 脚本** — 新增 `scripts/compute_lesson_activation.py`（PG→2aNzzy V/W）
|
||||
|
||||
**关键数据源映射：** 进线=2aNzzy C列日期 → 用户ID=2aNzzy → PG user_course_detail 首课日期 → 比对同天 → 写回 2aNzzy V/W
|
||||
|
||||
### 待办汇总
|
||||
|
||||
1. 📋 陈逸鸫确认行课转化改动方案 → 一口气改 6 处
|
||||
2. 📋 王虹茗 user_id 获取（需她发消息或陈逸鸫截图）
|
||||
3. 📋 数据转发王虹茗 + 写入大麦查询输出表 fd42b8
|
||||
4. 📋 销售看板 build 挂住问题排查
|
||||
5. 📋 Image2 生图任务执行
|
||||
6. ⏳ 全量 pipeline 聚光验证（子进程）
|
||||
|
||||
### 20:52 行课转化全量改动完成
|
||||
|
||||
[陈逸鸫确认] 行课记录新增指标：当日进线→当天行课 + 7天首课，6处改动已完成4处：
|
||||
|
||||
| # | 位置 | 改动 | 状态 |
|
||||
|---|------|------|------|
|
||||
| 1 | 2aNzzy | V/W/X 三列（首课日期/当日行课/7日内行课） | ✅ |
|
||||
| 2 | 3PRySY | AE-AH 四列（当日行课/当日行课率/7日内首课/7日内首课率）| ✅ |
|
||||
| 3 | C1HVN2 | 合并 三+五 →「线索→行课转化」| ✅ |
|
||||
| 4 | Base | 行课销售月(4)/5月漏斗(3)/销转销售月(2) 加字段 | ⚠️ bot权限不足，需手动 |
|
||||
| 5 | funnel-daily 看板 | 加行课转化指标 | 📋 待 build 脚本改 |
|
||||
| 6 | build 脚本 | compute_lesson_activation.py | ✅ |
|
||||
|
||||
**合并后的 C1HVN2 Section 三：**
|
||||
```
|
||||
三、线索→行课转化（5月27日）
|
||||
指标 小龙 吴迪 Bob Tom 日汇总
|
||||
企微新增 13 6 14 15 48
|
||||
当日行课 0 0 0 0 0
|
||||
当日行课率 0% 0% 0% 0% 0%
|
||||
7天线索→首课 1 2 2 2 10
|
||||
7天首课率 7.7% 33.3% 14.3% 13.3% 20.8%
|
||||
```
|
||||
|
||||
**5月全月数据（3PRySY 公式自动计算）：** Bob 当日行课率 7.4% > Tom 3.7% > 其余 0%；7天首课率 吴迪 28.6% > 小龙 14.3%
|
||||
|
||||
**关键实现细节：**
|
||||
- 2aNzzy V: PG chapter_settlement_data 最早日期
|
||||
- 2aNzzy W: 进线当天有 chapter 活动=1
|
||||
- 2aNzzy X: 进线 7 天内有 chapter 活动=1
|
||||
- 3PRySY 公式 =COUNTIFS(2aNzzy!O:O,month, 2aNzzy!W:W,1) 等
|
||||
- 440 条个体行课记录覆盖 3-5 月
|
||||
- compute_lesson_activation.py 进 cron 后每日自动刷新
|
||||
|
||||
**待办：** Base 字段需陈逸鸫在 UI 手动添加（行课销售月/5月漏斗/销转销售月各加 2-4 个数字字段）
|
||||
|
||||
|
||||
|
||||
## funnel-daily 看板行课激活指标 — 待修复 (21:49)
|
||||
|
||||
### Bug 1: `_read_activations` 未过滤汇总行
|
||||
`snapshot_funnel_daily.py` 读 2aNzzy O2:X4000 时，只检查 O/P 是否有值，没检查 Q (用户ID)，导致汇总行（无 user_id）也被计入。
|
||||
- **修复**: 加 `if not uid: continue`，只统计有 user_id 的个体小溪记录
|
||||
|
||||
### Bug 2: key 不匹配
|
||||
`collect_snapshot` 合计用 `"week_lesson"`，但 `_read_activations` 存的是 `"week"`，导致 合计 wk=0。
|
||||
- **修复**: 统一用 `"week"` key
|
||||
|
||||
### Bug 3: X列数据 O列分组失真
|
||||
X列 (7日内行课) 是基于 B列(进线日期) 计算的，但按 O列(行课月) 聚合会导致跨月用户重复计入。5月 O列=505行但个体记录只有约108条/月。
|
||||
- **初步方案**: 先按 O列+Q列过滤（过滤汇总行），暂时接受 O列分组。后续可考虑 B列分组。
|
||||
|
||||
### 待执行 (下次回话)
|
||||
```bash
|
||||
# 1. 修复 snapshot_funnel_daily.py 两处 bug
|
||||
# 2. 重建 snapshot + 看板
|
||||
cd /root/.openclaw/workspace-xiaoban/xhs-ark-dashboard
|
||||
source scripts/xiaoban_env.sh
|
||||
python3 scripts/snapshot_funnel_daily.py --date 2026-05-28
|
||||
python3 scripts/build_funnel_dashboard.py --month 5
|
||||
```
|
||||
|
||||
|
||||
### ~21:50 C1HVN2 日报格式修复
|
||||
|
||||
[陈逸鸫] 报 C1HVN2 Section 三「线索→行课转化」格式乱七八糟，数字是 text 而非数字格式。
|
||||
|
||||
**清理完成：**
|
||||
- G23:G25/G30 stray 值清除 ✅
|
||||
- Row 19（当日行课）/ Row 21（7天首课）→ int 型正常 ✅
|
||||
- Row 20/22（百分比行）→ 用 USER_ENTERED 写入 "0.0%"/"14.3%" 字符串 ✅
|
||||
- H 列测试残渣清除 ✅
|
||||
|
||||
**⚠️ Row 18（企微新增）仍是 text 格式** — 那些单元格有历史残留 TEXT 格式，底层值已是整数但飞书 UI 强制显示为文本。API 无法修改 cell numberFormat（`+batch-set-style` 是 stub，`repeatCell` 调飞书 API 返回 90204）。**需陈逸鸫在飞书里选中 B18:F18 → 右键「清除格式」手动修复。**
|
||||
|
||||
### ~21:55 lark-cli `+write` bug 修复
|
||||
|
||||
**根因：** `bin/lark-cli-impl` 的 `+write` action 从未读取 `--value-input-option` 参数，PUT URL 里也没加 `?valueInputOption=USER_ENTERED`。参数被完全忽略。
|
||||
|
||||
**修复：**
|
||||
```python
|
||||
vio = pargs.get("value_input_option", "RAW")
|
||||
url_params = f"?valueInputOption={vio}"
|
||||
result = api("PUT", f"{SHEETS_V2}/spreadsheets/{token}/values{url_params}", ...)
|
||||
```
|
||||
同理 `+append` 也加了参数支持。
|
||||
|
||||
**验证：** 修复后写 int 到全新单元格 → 飞书正确显示为数字 ✅。但 Row 18 旧单元格 TEXT 格式仍覆盖显示。
|
||||
|
||||
### ~22:00 飞书 Sheets API 公式求值测试
|
||||
|
||||
**结论：公式无法通过 API 写入并求值。** 即使 `valueInputOption=USER_ENTERED` 修复后，在全新单元格写 `=1+2` → FormattedValue 显示文本 "=1+2"，不计算为 3。
|
||||
|
||||
**飞书 Sheets API v2 的 USER_ENTERED 的作用：** 数字/日期解析正常 → 公式不支持求值。
|
||||
|
||||
**陈逸鸫提到 Cursor 可以写公式** — 待确认他用的是什么工具/API。可能路径：Cursor 用了不同端点（Sheets v3？）、不同 auth 方式、或者直接用 Feishu MCP。
|
||||
|
||||
### ~22:05 日报「公式映射」现状
|
||||
|
||||
**事实：** C1HVN2 Section 三当前数字是 `compute_lesson_activation.py` / `pipeline.py step 3` 写入的静态值，不是真正的 Feishu 公式。API 层无法写入公式。
|
||||
|
||||
**正确的「公式映射」方案：** 不强求单元格公式，而是 pipeline 脚本作为"公式"—从源表取数→计算→写入。Section 一/二的现有自动化就是这样运行的。需验证 pipeline step 3 是否已覆盖新增的当日行课/7天首课行，如果没覆盖就加进去。
|
||||
|
||||
### 当前 C1HVN2 待修复项
|
||||
|
||||
1. Row 18（企微新增）TEXT 格式 → 需陈逸鸫手动「清除格式」
|
||||
2. Section 四（销转情况）Row 26-29 数字也是 text 格式 → 同样需手动清除
|
||||
3. 当日行课/7天首课数据 → pipeline 脚本需覆盖 Section 三写入
|
||||
|
||||
30
memory/2026-05-29.md
Normal file
30
memory/2026-05-29.md
Normal file
@ -0,0 +1,30 @@
|
||||
# 2026-05-29 周五
|
||||
|
||||
## 上午 · 陈逸鸫
|
||||
|
||||
### Pipeline 状态确认
|
||||
- 三条工作日 cron 就绪（7:15/9:30/9:50），今天为首次自动执行日
|
||||
- 手动提前触发小溪行课 push + pull：1920 条已推送、782 条 ID 查询、30 条待小溪处理
|
||||
|
||||
### 成都 UID→销售修复
|
||||
- 修改 `pull_xiaoxi_results.py`：`load_chengdu_uid_owner()` 增加 phone→sales 兜底映射
|
||||
- 成都跳过从 38 降到 34（找回 4 条），仍有 34 条待排查
|
||||
|
||||
### ERCFsFo 403 写权限
|
||||
- `sync_master_xiaoxi.py` 写销转客户主表 `ERCFsFo4MhnF0ytGeCrc0Bb8n5f` 的"小溪学情"子表时报 403
|
||||
- CYFTsu（生产表）读写正常；ERCFsFo 能读不能写（user/bot 均 403）
|
||||
- 陈逸鸫说"之前都开了权限"，需进一步确认飞书 UI 中该表的协作权限
|
||||
|
||||
### 微伴今日更新
|
||||
- 收到微伴 xlsx，执行 `run_morning_weiban_from_dm.sh` 完成
|
||||
- 1a(43.5s) → 1a2(20.6s) → 2b(73.6s) → 4a(11.8s) 全部 OK
|
||||
- 日报 C1HVN2 一/三/四已刷新，二跳过（等聚光）
|
||||
- 聚光 1b+1c 等 9:30 cron
|
||||
|
||||
### 其他待办
|
||||
- xhs-tech-dashboard 需 clone URL（当前为 tar 解压，非 git 仓库）
|
||||
- Base 字段手动添加（bot 权限不足）
|
||||
- C1HVN2 Row 18 企微新增 TEXT 格式清除
|
||||
- funnel-daily 看板 snapshot 3 个 bug
|
||||
- L1-S1-U1《秘密基地》5 张投放底图
|
||||
- 漏斗看板发布方案（bot 不支持妙搭 html-publish）
|
||||
BIN
output/U1-L1-飞船打扫-map-v1.png
Normal file
BIN
output/U1-L1-飞船打扫-map-v1.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 5.5 MiB |
BIN
output/U1-L1-飞船打扫-map-v2.png
Normal file
BIN
output/U1-L1-飞船打扫-map-v2.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 6.0 MiB |
BIN
output/U1-L1-飞船打扫-map-v3.png
Normal file
BIN
output/U1-L1-飞船打扫-map-v3.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 6.2 MiB |
BIN
output/U1-L1-飞船打扫-map-v4.png
Normal file
BIN
output/U1-L1-飞船打扫-map-v4.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 707 KiB |
BIN
output/U1-L1-飞船打扫-map-v5.png
Normal file
BIN
output/U1-L1-飞船打扫-map-v5.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 2.6 MiB |
BIN
output/小龙订单_0421_0520.xlsx
Normal file
BIN
output/小龙订单_0421_0520.xlsx
Normal file
Binary file not shown.
196
scripts/build_dragon_order_xlsx.py
Normal file
196
scripts/build_dragon_order_xlsx.py
Normal file
@ -0,0 +1,196 @@
|
||||
#!/usr/bin/env python3
|
||||
"""查询小龙名下 4/21-5/20 订单详情（从MySQL反查用户ID）"""
|
||||
import json, urllib.request, re, sys, os, subprocess
|
||||
from openpyxl import Workbook
|
||||
from openpyxl.styles import Font, PatternFill, Alignment, Border, Side
|
||||
from openpyxl.utils import get_column_letter
|
||||
|
||||
APP_ID = 'cli_a92fc074fb5edcb5'
|
||||
APP_SECRET = 'jzQ8UoNb06rX8147V52icdWF7XN8Su2K'
|
||||
SPREADSHEET_TOKEN = 'ERCFsFo4MhnF0ytGeCrc0Bb8n5f'
|
||||
|
||||
def get_token():
|
||||
url = 'https://open.feishu.cn/open-apis/auth/v3/tenant_access_token/internal'
|
||||
data = json.dumps({'app_id': APP_ID, 'app_secret': APP_SECRET}).encode()
|
||||
req = urllib.request.Request(url, data=data, headers={'Content-Type': 'application/json'})
|
||||
with urllib.request.urlopen(req) as resp:
|
||||
return json.loads(resp.read())['tenant_access_token']
|
||||
|
||||
def read_sheet(token, sheet_id, cols_range):
|
||||
url = f'https://open.feishu.cn/open-apis/sheets/v2/spreadsheets/{SPREADSHEET_TOKEN}/values/{sheet_id}!{cols_range}?valueRenderOption=ToString&dateTimeRenderOption=FormattedString'
|
||||
req = urllib.request.Request(url, headers={'Authorization': f'Bearer {token}', 'Content-Type': 'application/json'})
|
||||
with urllib.request.urlopen(req) as resp:
|
||||
result = json.loads(resp.read())
|
||||
if result.get('code') != 0: return None
|
||||
return result['data']['valueRange']['values']
|
||||
|
||||
def parse_date(s):
|
||||
if not s: return None
|
||||
s = str(s).strip()
|
||||
m = re.match(r'(\d+)月(\d+)日', s)
|
||||
if m: return (int(m.group(1)), int(m.group(2)))
|
||||
return None
|
||||
|
||||
def lookup_uids_by_phone(phones):
|
||||
"""Query MySQL vala_app_account by masked phone patterns."""
|
||||
patterns = list({f'{p[:3]}****{p[-4:]}' for p in phones if p})
|
||||
if not patterns: return {}
|
||||
|
||||
conds = " OR ".join([f"tel='{p}'" for p in patterns])
|
||||
sql = f"""mysql -h bj-cdb-dh2fkqa0.sql.tencentcdb.com -P 27751 -u read_only -p'fsdo45ijfmfmuu77$%^&' -e "USE vala_user; SELECT id, tel FROM vala_app_account WHERE {conds};" 2>/dev/null"""
|
||||
|
||||
result = subprocess.run(sql, shell=True, capture_output=True, text=True)
|
||||
lines = result.stdout.strip().split('\n')
|
||||
|
||||
# Build: masked_phone -> list of account IDs
|
||||
masked = {}
|
||||
for line in lines[1:]:
|
||||
parts = line.split('\t')
|
||||
if len(parts) >= 2:
|
||||
masked.setdefault(parts[1], []).append(parts[0])
|
||||
|
||||
# Map: full_phone -> first account ID
|
||||
uid_map = {}
|
||||
for p in phones:
|
||||
if not p: continue
|
||||
pat = f'{p[:3]}****{p[-4:]}'
|
||||
ids = masked.get(pat, [])
|
||||
uid_map[p] = ids[0] if ids else ''
|
||||
return uid_map
|
||||
|
||||
# Step 1: Get orders
|
||||
print("Reading orders...", file=sys.stderr)
|
||||
token = get_token()
|
||||
order_rows = read_sheet(token, 'e12fb9', 'A:N')
|
||||
|
||||
matched = []
|
||||
for row in order_rows:
|
||||
if len(row) < 3: continue
|
||||
if str(row[0]).strip() != '小龙': continue
|
||||
d = parse_date(row[2])
|
||||
if d and (4,21) <= d <= (5,20):
|
||||
vals = [str(c) if c else '' for c in row]
|
||||
while len(vals) < 14: vals.append('')
|
||||
matched.append({'nickname': vals[3], 'order_date': vals[2], 'vals': vals})
|
||||
|
||||
# Step 2: Get phones from fill sheet
|
||||
fill_rows = read_sheet(token, '1RFMqc', 'A:I')
|
||||
phone_by_key = {}
|
||||
for row in fill_rows:
|
||||
if len(row) < 3: continue
|
||||
if str(row[0]).strip() != '小龙': continue
|
||||
nickname = str(row[3]).strip() if len(row) > 3 and row[3] else ''
|
||||
order_date = str(row[2]).strip() if row[2] else ''
|
||||
phone = str(row[4]).strip() if len(row) > 4 and row[4] else ''
|
||||
key = (nickname, order_date)
|
||||
if key[1] and key not in phone_by_key and phone:
|
||||
phone_by_key[key] = phone
|
||||
|
||||
# Step 3: Lookup UIDs from MySQL
|
||||
all_phones = [phone_by_key.get((o['nickname'], o['order_date']), '') for o in matched]
|
||||
uid_map = lookup_uids_by_phone(all_phones)
|
||||
print(f"UIDs found: {sum(1 for v in uid_map.values() if v)}/{len(uid_map)}", file=sys.stderr)
|
||||
|
||||
# Step 4: Build results
|
||||
results = []
|
||||
for o in matched:
|
||||
v = o['vals']
|
||||
phone = phone_by_key.get((o['nickname'], o['order_date']), '')
|
||||
uid = uid_map.get(phone, '')
|
||||
results.append({
|
||||
'nickname': v[3], 'phone': phone, 'uid': uid,
|
||||
'online_date': v[1], 'order_date': v[2], 'product': v[5],
|
||||
'amount': float(v[6]) if v[6] else 0,
|
||||
'channel': v[4], 'attribution': v[9],
|
||||
'status': '有效成单' if v[11]=='1' else ('全额退' if v[11]=='0' else v[10]),
|
||||
'valid': v[11]=='1'
|
||||
})
|
||||
|
||||
missing_uid = sum(1 for r in results if not r['uid'])
|
||||
print(f"With UID: {len(results)-missing_uid}/{len(results)}", file=sys.stderr)
|
||||
|
||||
# Step 5: Build Excel
|
||||
wb = Workbook()
|
||||
ws = wb.active
|
||||
ws.title = "小龙订单明细"
|
||||
|
||||
hdr_font = Font(name='微软雅黑', bold=True, size=11, color='FFFFFF')
|
||||
hdr_fill = PatternFill(start_color='2F5496', end_color='2F5496', fill_type='solid')
|
||||
hdr_align = Alignment(horizontal='center', vertical='center', wrap_text=True)
|
||||
cell_font = Font(name='微软雅黑', size=10)
|
||||
cell_align = Alignment(horizontal='center', vertical='center')
|
||||
money_font = Font(name='微软雅黑', size=10, bold=True)
|
||||
thin_border = Border(left=Side(style='thin', color='D9D9D9'), right=Side(style='thin', color='D9D9D9'),
|
||||
top=Side(style='thin', color='D9D9D9'), bottom=Side(style='thin', color='D9D9D9'))
|
||||
zebra_fill = PatternFill(start_color='F2F7FB', end_color='F2F7FB', fill_type='solid')
|
||||
sum_fill = PatternFill(start_color='FFF2CC', end_color='FFF2CC', fill_type='solid')
|
||||
sum_font = Font(name='微软雅黑', bold=True, size=10)
|
||||
db_fill = PatternFill(start_color='E8F5E9', end_color='E8F5E9', fill_type='solid')
|
||||
db_font = Font(name='微软雅黑', size=9, color='2E7D32', italic=True)
|
||||
|
||||
headers = ['序号', '微信昵称', '手机号', '用户ID', '进线日期', '落单日期', '产品', '实付金额', '成交渠道', '渠道归属', '订单状态']
|
||||
|
||||
# Title
|
||||
ws.merge_cells('A1:K1')
|
||||
ws['A1'].value = '小龙名下订单明细（落单时间：2026-04-21 ~ 2026-05-20）'
|
||||
ws['A1'].font = Font(name='微软雅黑', bold=True, size=14, color='2F5496')
|
||||
ws['A1'].alignment = Alignment(horizontal='center', vertical='center')
|
||||
ws.row_dimensions[1].height = 36
|
||||
|
||||
# Summary
|
||||
total_rev = sum(r['amount'] for r in results)
|
||||
sx = sum(1 for r in results if r['attribution']=='销转')
|
||||
dr = sum(1 for r in results if r['attribution']=='达人')
|
||||
dn = sum(1 for r in results if r['attribution']=='端内')
|
||||
ws.merge_cells('A2:K2')
|
||||
ws['A2'].value = f'共 {len(results)} 单 | 全部有效 | 总额 ¥{total_rev:,.0f} | 销转{sx} / 达人{dr} / 端内{dn} | 用户ID: 已匹配{len(results)-missing_uid}/{len(results)} (MySQL反查)'
|
||||
ws['A2'].font = Font(name='微软雅黑', size=9, color='666666')
|
||||
ws['A2'].alignment = Alignment(horizontal='center', vertical='center')
|
||||
ws.row_dimensions[2].height = 24
|
||||
|
||||
# Header
|
||||
for ci, h in enumerate(headers, 1):
|
||||
c = ws.cell(row=3, column=ci, value=h)
|
||||
c.font = hdr_font; c.fill = hdr_fill; c.alignment = hdr_align; c.border = thin_border
|
||||
ws.row_dimensions[3].height = 28
|
||||
|
||||
# Data
|
||||
for idx, r in enumerate(results):
|
||||
rn = idx + 4
|
||||
data = [idx+1, r['nickname'], r['phone'], r['uid'], r['online_date'],
|
||||
r['order_date'], r['product'], r['amount'], r['channel'],
|
||||
r['attribution'], r['status']]
|
||||
for ci, val in enumerate(data, 1):
|
||||
c = ws.cell(row=rn, column=ci, value=val)
|
||||
c.font = cell_font; c.alignment = cell_align; c.border = thin_border
|
||||
if ci == 4: # UID column - highlight DB-sourced
|
||||
c.font = db_font if r['uid'] else Font(name='微软雅黑', size=9, color='999999', italic=True)
|
||||
c.fill = db_fill if r['uid'] else PatternFill(start_color='FFF0F0', end_color='FFF0F0', fill_type='solid')
|
||||
if not r['uid']: c.value = '未匹配'
|
||||
elif idx % 2 == 0:
|
||||
c.fill = zebra_fill
|
||||
# Money format
|
||||
mc = ws.cell(row=rn, column=8)
|
||||
mc.number_format = '#,##0'; mc.font = money_font
|
||||
|
||||
# Summary row
|
||||
sr = len(results) + 4
|
||||
ws.merge_cells(f'A{sr}:G{sr}')
|
||||
for c in range(1, 12):
|
||||
ws.cell(row=sr, column=c).fill = sum_fill
|
||||
ws.cell(row=sr, column=c).border = thin_border
|
||||
ws.cell(row=sr, column=c).font = sum_font
|
||||
ws.cell(row=sr, column=1, value='合计').alignment = Alignment(horizontal='right', vertical='center')
|
||||
tc = ws.cell(row=sr, column=8, value=total_rev)
|
||||
tc.number_format = '#,##0'; tc.alignment = cell_align
|
||||
|
||||
# Col widths
|
||||
for i, w in enumerate([6, 22, 16, 10, 10, 10, 14, 12, 16, 10, 12], 1):
|
||||
ws.column_dimensions[get_column_letter(i)].width = w
|
||||
ws.freeze_panes = 'A4'
|
||||
ws.auto_filter.ref = f'A3:K{sr}'
|
||||
|
||||
out = '/root/.openclaw/workspace-xiaoban/output/小龙订单_0421_0520.xlsx'
|
||||
wb.save(out)
|
||||
print(f"Done: {out}", file=sys.stderr)
|
||||
print(out)
|
||||
119
scripts/query_order_dragon.py
Normal file
119
scripts/query_order_dragon.py
Normal file
@ -0,0 +1,119 @@
|
||||
#!/usr/bin/env python3
|
||||
"""查询小龙名下 4/21-5/20 落单的订单详情（含手机号和用户ID）"""
|
||||
|
||||
import json, urllib.request, re, sys
|
||||
|
||||
APP_ID = 'cli_a92fc074fb5edcb5'
|
||||
APP_SECRET = 'jzQ8UoNb06rX8147V52icdWF7XN8Su2K'
|
||||
SPREADSHEET_TOKEN = 'ERCFsFo4MhnF0ytGeCrc0Bb8n5f'
|
||||
SHEET_FILL = '1RFMqc' # 全量销售填表（含手机号/用户ID）
|
||||
SHEET_ORDER = 'e12fb9' # 订单汇总（仅成单字段）
|
||||
|
||||
def get_token():
|
||||
url = 'https://open.feishu.cn/open-apis/auth/v3/tenant_access_token/internal'
|
||||
data = json.dumps({'app_id': APP_ID, 'app_secret': APP_SECRET}).encode()
|
||||
req = urllib.request.Request(url, data=data, headers={'Content-Type': 'application/json'})
|
||||
with urllib.request.urlopen(req) as resp:
|
||||
return json.loads(resp.read())['tenant_access_token']
|
||||
|
||||
def read_sheet(token, sheet_id, cols_range):
|
||||
url = f'https://open.feishu.cn/open-apis/sheets/v2/spreadsheets/{SPREADSHEET_TOKEN}/values/{sheet_id}!{cols_range}?valueRenderOption=ToString&dateTimeRenderOption=FormattedString'
|
||||
req = urllib.request.Request(url, headers={
|
||||
'Authorization': f'Bearer {token}',
|
||||
'Content-Type': 'application/json',
|
||||
})
|
||||
with urllib.request.urlopen(req) as resp:
|
||||
result = json.loads(resp.read())
|
||||
if result.get('code') != 0:
|
||||
return None
|
||||
return result['data'].get('valueRange', {}).get('values', [])
|
||||
|
||||
def parse_date(s):
|
||||
if not s:
|
||||
return None
|
||||
s = str(s).strip()
|
||||
m = re.match(r'(\d+)月(\d+)日', s)
|
||||
if m:
|
||||
return (int(m.group(1)), int(m.group(2)))
|
||||
m = re.match(r'(\d{4})[-/](\d{1,2})[-/](\d{1,2})', s)
|
||||
if m:
|
||||
return (int(m.group(2)), int(m.group(3)))
|
||||
return None
|
||||
|
||||
START, END = (4, 21), (5, 20)
|
||||
|
||||
token = get_token()
|
||||
|
||||
# Step 1: Read order summary to get the list of matched orders (for row ordering)
|
||||
print("Step 1: Reading order summary...", file=sys.stderr)
|
||||
order_rows = read_sheet(token, SHEET_ORDER, 'A:N')
|
||||
matched_orders = []
|
||||
for i, row in enumerate(order_rows):
|
||||
if len(row) < 3:
|
||||
continue
|
||||
owner = str(row[0]).strip() if row[0] else ''
|
||||
if owner != '小龙':
|
||||
continue
|
||||
d = parse_date(row[2])
|
||||
if d and START <= d <= END:
|
||||
vals = [str(c) if c else '' for c in row]
|
||||
while len(vals) < 14:
|
||||
vals.append('')
|
||||
matched_orders.append({
|
||||
'row': i+1, 'nickname': vals[3], 'order_date': vals[2],
|
||||
'vals': vals
|
||||
})
|
||||
|
||||
print(f"Orders: {len(matched_orders)}", file=sys.stderr)
|
||||
|
||||
# Step 2: Read full fill sheet for 小龙 to get phone and UID
|
||||
print("Step 2: Reading fill sheet for phones/UIDs...", file=sys.stderr)
|
||||
fill_rows = read_sheet(token, SHEET_FILL, 'A:M') # A-M: 销售归属~退款金额
|
||||
|
||||
# Build lookup: (nickname, order_date) -> (phone, uid)
|
||||
fill_lookup = {}
|
||||
for row in fill_rows:
|
||||
if len(row) < 3:
|
||||
continue
|
||||
owner = str(row[0]).strip() if row[0] else ''
|
||||
if owner != '小龙':
|
||||
continue
|
||||
nickname = str(row[3]).strip() if len(row) > 3 and row[3] else ''
|
||||
order_date = str(row[2]).strip() if row[2] else ''
|
||||
phone = str(row[4]).strip() if len(row) > 4 and row[4] else ''
|
||||
uid = str(row[8]).strip() if len(row) > 8 and row[8] else ''
|
||||
key = (nickname, order_date)
|
||||
if key[1] and (key[1] not in fill_lookup or (phone or uid)):
|
||||
fill_lookup[key] = (phone, uid)
|
||||
|
||||
# Step 3: Merge and output
|
||||
print("Step 3: Merging...", file=sys.stderr)
|
||||
|
||||
# Stats
|
||||
total_revenue = 0
|
||||
valid_count = 0
|
||||
|
||||
print(f"## 小龙名下订单详情（落单：2026-04-21 ~ 2026-05-20）\n")
|
||||
print(f"共 {len(matched_orders)} 条 · 数据来源：销转客户主表\n")
|
||||
|
||||
for idx, o in enumerate(matched_orders, 1):
|
||||
v = o['vals']
|
||||
key = (o['nickname'], o['order_date'])
|
||||
phone, uid = fill_lookup.get(key, ('', ''))
|
||||
|
||||
amount = float(v[6]) if v[6] else 0
|
||||
is_valid = "✅" if v[11] == '1' else ("⚠️退" if v[11] == '0' else v[11])
|
||||
|
||||
total_revenue += amount
|
||||
if v[11] == '1':
|
||||
valid_count += 1
|
||||
|
||||
phone_str = f"📱{phone}" if phone else "📱—"
|
||||
uid_str = f"🆔{uid}" if uid else "🆔—"
|
||||
|
||||
amt_str = f"{int(float(v[6])):,}" if v[6] else "0"
|
||||
print(f"{idx}. {v[3]} {phone_str} {uid_str}")
|
||||
print(f" 进线 {v[1]} 落单 {v[2]} {v[5]} ¥{amt_str} {v[4]} · {v[9]} {is_valid}")
|
||||
|
||||
print()
|
||||
print(f"**汇总：{len(matched_orders)} 单 | 有效 {valid_count} 单 | 总额 ¥{total_revenue:,.0f}**")
|
||||
@ -1 +1 @@
|
||||
Subproject commit d7e43830b634a87c3b13c65f937ca77a3aa2d22d
|
||||
Subproject commit 4a6f6d3443c1d303fc54e82ddf55ffc74f4ea2c5
|
||||
Loading…
Reference in New Issue
Block a user